aero307collection

[aero307collection]

Reflection and Additional Thoughts

I feel like that I understand flow separation and stall a lot better now since the start of 307 because the testing we did in this class really allowed to me see physically what was happening when a wing stalls and when the flow separates from the wing.  In addition, I was able to see how flow can be reattached through the use of tufts, which was really interesting to me. I can’t really think of an aerodynamics concept that confused me, but something that trips me up a lot is using MatLab to creation functions and use that function to calculate the data we get from the wind tunnel. Even though I have taken a lot of MatLab courses I still sometimes feel confused on how to implement the data we found and convert it to useful data such as Cp, Cd, and drag numbers. The highlight of this course would definitely have to be the mini project portion of the class because I feel like what we did with the projects is what its like to work in industry. We are given a problem and our team has to figure out the best way to tackle the problem with some guidance and pointers from someone that has more knowledge than us. It’s really hard to pick a lowlight in this class because I really enjoyed every lab day we have, but if I had to choose one then it would be the force balance lab only because it was similar to what we did before and the overall lab was really simple and straight forward. I actually really enjoyed aerodynamics and I plan on taking your CFD class in the future. As for the future, I’m really interested in working with/on drones and UAVs because I think they are really interesting and potentially a new way to transport objects without the use of a physical pilot on board. In addition, I read that a lot of companies interested in small aircraft and drones to help with their operations; some of these companies include Uber for their “sky-taxis”, medical companies which use drones to take blood or medicine to areas that are far from the main city, Amazon for their delivery system, and many more companies. I believe that this will be the future of aeronautics and I would love to help work on any of these projects in future.

 Lastly I would like to thank you Dr. Doig from a great quarter. I really enjoyed your lab and the way you allow us to test, run and work on experiments in the wind tunnel and water tunnel. It really allowed me to get hands on experience working on real life projects and I’m grateful for that. If possible, I would love to help you on any projects you might have this summer since I will be in SLO for the summer. I really want to keep learning and working on projects in order to implement what I’ve learned for the past three years. My email is nguyenle@calpoly.edu and I hope to hear from you soon! Have a great summer Dr. Doig!

[aero307collection]

Active Flow Control Project Finale and Presentations

On Monday of the tenth week, my group and I continued to test the wing tip jet velocity in order to find the fastest jet stream. The data table located below shows the results we found from our testing on the second day. It can be seen that the testing varied a lot from the first day because the fastest velocity we got from the wing tip jet was 2.8 m/s at a frequency of 100 Hz and a voltage of 100V. However, on the second day of testing we got a wing tip jet velocity of 5.9 m/s at the same frequency and voltage. One of the reasons why we believe there was such a big change was because it was really difficult to measure the exact spot from which the air came out of the wing tip jet due to such as small slit. The sweet spot that we found by using the hot wire anemometer was only about 1mm.

Figure 1: Data from wing tip jet velocities (Day 1) 

Figure 2: Data from wing tip jet velocities (Day 2)

Afterwards, we decided to put the wing into the wind tunnel again to test the wing tip jet at it’s highest velocity. We started the wind tunnel up to 15 m/s and then dropped it down to around 10 m/s so that we could see how the wing tip jets would affect the tufts on the wing. The reason we had to start the wind tunnel up at 15 m/s is because when we ran the tunnel at 10 m/s the tuft did not move at all. This is similar to the hysteresis affect that we learned in Aero 302. When the wing tip jet was turned on we found from our videos that the frequency and amplitude of the tuft’s oscillation was different with the speaker’s on, however we are unsure of how that correlates to drag. From the pictures below it can be seen that the tufts did decrease in its amplitude and from our analysis when the wing tip jets was turned off the tuft frequency was 7.74 Hz. When the wing tip jets was turned on the tuft frequency was 6.86 Hz.

Figure 3: Different in tuft amplitude (Left = Speaker off, Right = Speaker on)

The overall results of our active flow control project was that we were able to sand the leading edge of the airfoil to make it a lot smoother than before and there is little to no lip on the leading edge. Another thing we did was we found the optimal speaker setting, 100 Hz and 100V, in order to get the fastest wing tip jet velocity. Lastly, we were able to visually confirm that there was a lower amplitude and lower frequency oscillation of trailing edge wing tip tuft when the tip jet was used.

As for future recommendations and possible things to do, I would say that if anyone wanted to test the pressure ports on the wing again then they would need to clear the pressure ports of the sanded acrylic debris. Another thing would to develop a way to repeatedly and accurately measure the tip jet velocities. Lastly, develop a way to quantify the effect the tip jet has on the wing tip vortex such as wake survey or testing the pressure ports.

On the second day of lab, we presented our presentation to the class. I thought our presentation went really well. I love the flow visualization done by the Tesla car team and their use of the UV lights and dies. It was one of the coolest flow viz I’ve seen and it was interesting that they found some interesting results from their test.

[aero307collection]

Active Flow Control Project: Wind Tunnel Test and Jet Velocity Measurement

On Monday of the ninth week our group did not meet because of Memorial Day. 

On the second day of lab, Dr. Doig returned and he was fine with the way we were sanding the leading edge so my group and I continued to sand the leading edge for a bit. Since we were low on time due to not having lab on Monday, we decided to take a pause on sanding the wing and test the wing in the wind tunnel. The first thing we had to do was mount the wing on the traverse, which took a bit of time because we had to unscrew the side of the wind tunnel to fit the wing in as well as accurately mount the wing so that it would fit through the gap. Next, Dr. Doig and our group took a bit of time trying to figure out how to get the traverse to move since the machine was acting up. Eventually, we got the traverse to work and it can be seen below that the wing had a tight fit into the open panel in the test section.

Figure 1: Mounting the wing on the traverse

Figure 2: Fitting the wing into the test section gap

For our testing we ran the wind tunnel at 15 m/s and used smoke in order to see if there was an effect from the tip jets on the wing tip vortex. In the first video, it can be seen that the smoke did not help in seeing anything at all. Also there was no noticeable different between having the jet turned on or off. Later on, we figured that we would need the wing tip jet velocity to better match the Mach tunnel speed for there to be any effect. On the other hand, the tuft on the wing tip’s trailing edge did appear to swirl with the vortex that was generated.

Figure 3: Day 1 wing test with fog 

Since we saw that were was no difference in the vortex around the wing tip, we decided to investigate the performance of the tip jet out of an air flow. To do this, we measured the tip jet exit velocity at various combinations of voltages and frequencies. In the data table below, we found that the jet velocity increased as the voltage was constant and the frequency increased. However, there was a point in which increasing the voltage did not really increase the jet velocity that much. This was really interesting to us because we were finding something that the thesis did not stated at all and the data could help with further experiments in the future.

Figure 4: Day 1 test results for tip jet velocities

During the process of gathering results, we found out the the waveform generator was set in the wrong settings when we tested in the wind tunnel so that tests we did might have been incorrect. We were supposed to set the frequency to square waves, however during the wind tunnel test we had the frequency set to sine waves. This might have an effect on either we see the performance of the wing tip jet, however I don’t believe that the change will effect it too much. We will test again next time and make sure that the settings are correct. From our results, we found that the fastest wing tip velocity was 3 m/s and if we paired that with the 15 m/s free stream flow from the wind tunnel, the resultant vector of the two velocity components is not much different than the free stream velocity component. Therefore, it seems that the wing tip jet won’t really affect the flow. After recording all the data, we were out of time for the lab period, but we decided to repeat test again in order to find the fastest tip jet combination for the next lab.

Figure 5: Measuring the wing tip velocities using a anemometer 

[aero307collection]

Active Flow Control Project: Mounting and Sanding

On Monday of the eighth week of lab, my lab groups and I worked on making a sanding block to sand the leading edge of the active flow control wing as well as finding a way to mount the wing to the wing tunnel. Josh was responsible for designing the sanding block using Solidworks and then we would print the block using a 3D printer that Dr. Doig had. Olivia and I had the task of finding away to amount the wing on the traverse so that it can be placed horizontally in the wind tunnel.

Below is the Solidworks drawing that Josh made so that we can 3D our sanding block. After talking with Dr. Doig we determined that our number one priority was to sand down the leading edge of the active flow control wing because when the graduate student manufactured the wing there was a steam created from the upper and lower parts of the wing. Our hope is to sand down the leading edge so that there is a smooth surface along the leading edge. The smooth surface will produce more accurate results, which made it our first priority. After Josh designed the sanding block, we went into the 3D printing room and began to 3D print the part. I never 3D printed anything before so it was really interesting to see how the machine worked and all the different options there where to print different parts. Since the 3D printed part took awhile, we had to leave it to run past the time we had for class, but Dr. Doig said he would watch over it and that we could grab it on Wednesday.

Figure 1: Solidworks model of our sanding block

While Josh was designing the sanding block, Olivia and I tried to find away to mount the wing onto the traverse so that it can be placed into the wind tunnel. The first thing I did was unscrew and took off the wood panel that was attached to the wind tunnel so that there would be a small opening for the wing to go into. After that I unscrewed the base attachment of the wing so that Olivia and I could find a way to mount the base onto the traverse. When we tried to mount the base to the traverse only two bottom holes of the base could a line to the traverse so we decided to drill to more holes in order to keep the base of the wing stable. We drew holes that would a line with the holes of the traverse and Olivia when over into the IME lab to fiind someone to drill the two holes for us. After she came back with the base, we mounted the base onto the traverse using only three out of the four holes because we could not find anymore screws and nuts that would fit into the holes that we had. Fortunately, the three screws were strong enough to hole the base and wing up. At the end of our lab we tried to move the traverse with the wing attached to it so that we could see if it will fit inside the wing tunnel except we didn’t know how to run the traverse. Luckily, Kyle was there to try and help us, however he could not figure out why the traverse wasn’t working so we did not get to see if the wing was able to fit in the wind tunnel yet.

On the second day of lab, we found our 3D printed sanding block which looked like it could do the job very nicely. We started with a very gritty piece of sand paper so that we could sand the leading edge, however we found that the sand paper was mostly just scratching the wing and it did not benefit at all. Afterwards, we tried a different approach in which we used a finer grain of sand paper and we just tore a little strip of it so that it would fit on the bottom curve of the sanding block. Sanding the leading edge of the wing was a lot better with the second approach. We decided to just sand the part of the wing that was closest to the base all the way up until we got to the first pressure port because we wanted to make sure the sanding was done properly before we attempt to sand the part of the wing where the pressure ports were at. In addition, we didn’t want any scratches to be by the pressure ports because that would act similar to trip strips and result in inaccurate data. Once we have lab again and Dr. Doig comes back we will talk to him about our sanding process and see if the results are okay to continue on the part of the wing with the pressure ports.

Figure 2: Sanding down the leading edge of the wing

[aero307collection]

Week 7: 4412 Force Balance testing and New Project!

One Monday of the seventh week of lab, my group and I tested both the wings in the wind tunnel and collected data. Our first test was running the wind tunnel with just the sting and the mount of the wing to get the initial force data so that we can use it later when we collect the data from the wing. Afterwards, we ran both the wings at an angle attack of 0, 5, 10, 12, 14, 16, 18, and 20 degrees. We only decided to test at one speed for all the angles of attack, which was 20 m/s. That equates to a Reynolds number of around 154,500.

Figure 1: Cl vs Alpha plot from our collected data

From the graph it can be seen that the wing with the higher aspect ratio of 5.797 stalled before the wing with the lower aspect ratio of 3.0404. I think this happened because the down wash from the wingtip vortices are delaying the stall due to the flow being attached longer over the wing at higher angles of attack. There seems to be a delay in stall of about 4 degrees. Below is a plot from XFLR5 comparing the coefficient of lift vs the angle of attack (alpha). From the plot it seems that both the wings begin to stall at 14 degrees, which isn’t really true because from our data it shows that the higher aspect ratio tends to stall earlier. The XFLR5 plots might have an error in them so we will check the data from it when we finish putting together our report.

Figure 2: Cl vs Alpha plot from XFLR5 (Blue line=High AR, Yellow line=Low AR)

The next plot we made was comparing the coefficient of drag vs. alpha. The plots for both wings seem to be positively linear which makes sense because the higher the pitch angle, the more drag there will be due to there being more area of the wing against the flow of the air. The plot indicates that the higher aspect ratio wing has more drag than the lower aspect ratio and I think this is because the higher aspect ratio wing has more frontal area than the lower aspect ratio thing, therefore the drag will be more on objects that are bigger.

Figure 3: Cd vs Alpha plot from out collected data

Lastly, the plot done by XFLR below has the same shape as the one done in MatLab, but again the values do not seem to line up with our data. In addition, lower aspect ratio drag coefficient seems to be a little bit bigger than the higher aspect ration and that does not relate to the data that we got from MatLab. Again we will try to fix the XFLR5 data before we turn in our report. The settings for XFLR5 is a bit different since I used XFOIL beforehand, but I will try my best to match our data to the best of my ability.

Figure 4: Cd vs Alpha plot from XFLR5 (Blue line=High AR, Yellow line=Low AR)

On the second day of lab, I was placed into the same group as the 4412 Force Balance lab, which was great because we could jump right into our project. The project that we got assigned to was the active flow control project. I was really excited to get the opportunity to work on this because the idea of the wing sounds really interesting. For most of the lab time, my group and I read over the thesis from the graduate students that worked on it in order for us to get familiar with the wing and its design concept. After we read through the thesis we developed a test plan in order to plan out what we wanted to do for the next three weeks. Some things in our test plan included: sanding down the rough leading edge, making sure the pressure ports are identified correctly, seeing if the pressure ports are reading correct data, and experimenting with the tip jet function of the wing. Since we had a bit of extra time after we did our test plan, my group and I decided to test out the speakers inside the wing to determine of it was in working condition. To our surprise, the speakers in the wing worked perfectly fine! We also identified the harmful frequency of the speakers to be at 290 Hz. Next week we hope to sand down the leading edge and clean out the pressure ports before we run it in the wind tunnel.

[aero307collection]

Week 6 - Flow Viz Presentation and Prep for Force Balance Lab

On Monday of the sixth week of lab, we had presentations on all of the flow viz challenge labs. I thought that our presentation went really well, except we did have a lot of information on our slides. In the future, I will make sure to not have too much text on the slide because it will overwhelm the people watching the presentation. I think we did a good job with the material we presented and I thought the overall lab was a big success. I really liked how open ended the lab was in that we were able to pick whatever we wanted to do a flow viz on and then execute the way we wanted to. I feel that this approach is what it’s like in the real world where we have to create our own lab and find different ways to test whatever we are testing. Another thing I did was observe the various presentations that were given during our lab time. The toy truck and dye flow viz really caught my eye because I thought the overall presentation and the delivery was done really well. In addition, the lab seems really applicable to real life trucks, except for the fact that the windows were not fully cut out. Another lab I found interesting was the IR lab because I didn’t know very much about IR to begin with, so seeing how it was used and the videos was super cool.

On the second day of lab, my new group mates and I did not test since we were Group 3 so we worked on grabbing the data that Kyle sent us and put it into MatLab. The hardest part of implementing the data, which all the groups had in the conference room was that we had to make a code that would read in the data from each of the pages in Excel. Along with that we only read in two of the columns since the other column was the data and time which each run occurred at. Josh was mainly responsible for creating the code. While he was doing that I started to work on getting the theoretical data for the 4412 wing using XFOIL. After trying to use XFOIL for awhile and not being able do get much useful data; I switched to XFLR5. I took a while to learn XFLR5 because it’s a bit different than XFOIL, but there seems to be many more features. I watched a couple YouTube videos to learn about the program and I was able to create a Cl vs alpha graph similar to the one Josh made with the test data Kyle sent us.

 Figure 1: Graph of Cl vs alpha from the test data given

Figure 2: Graph of 4412 wing Cl vs. Alpha at ~15m/s and ~20m/s

From the graph it can be seen that the two graphs are pretty similar in the coefficient of lift vs the angle of attack. The amount of lift seems normal. For our test on Monday, we will be testing at two different speeds, 15m/s and 20m/s which we will calculate the Reynolds number for. The angle attack we will be testing at will be 0, 5, 10, and 15 degrees. The testing shouldn’t be hard at all, but I’m hoping we can get some good data so that we can get some accurate graphs!

[aero307collection]

Week 5 - Flow Viz Bridge Testing

During the fifth week of our lab my flow viz group and I tested our cable-stay bridge in the wind tunnel. Our goal of the lab was to observe flow viz, calculate Reynolds Number, calculate vortex shedding, calculate Strouhal Number, and determine the bridge and cable oscillation frequencies to compare to the Tacoma Narrows Bridge. The first thing we did when we went to the wind tunnel was mount the bridge perpendicular to the air flow so that we can see the how the air flows around the bridge.

Figure 1: Initial bridge set up in the wind tunnel 

It can be seen from the picture above that the bridge taped down using aluminum tape and it is perpendicular to the airflow. In addition, we wanted to add some precaution to prevent the bridge from flying down the tunnel if the taped failed so we put two 15lbs weights on each side of the bridge to prevent it from moving. As I will talk about later on, the initial setup above did not work because the bridge deck would fly upwards due to the trapped airflow underneath the bridge. Next, we started up the fog machine and turned on the fog light above the wind tunnel so that we can see the vortex shedding from the bridge. After, the initial setup we started the wind tunnel at around 100 RPM (~5.5m/s) and we could see some vortex shedding start to show, however the bridge was very still and did not oscillate at all. In order to try to get it to oscillate, we graduate went up by 20 RPMs until we reached 200 RPM (~11m/s). At 200 RPM was when we saw our first flaw in our initial setup.

Figure 2: Initial setup failure 

The picture above is what happened when we reached 200 RPM. This was caused by the airflow getting caught underneath our bridge and also the fact that our bridge was not 100 percent flat where the deck is. In order to fix this, we taped the deck to the pylons in order to keep the deck from moving. We tested again up to 250RPM, but we did not see any oscillation at all. We concluded that by taping the deck to the pylon, the deck had very little space to oscillate and the cardboard deck was pretty strong in the middle so it would not bend. The last change we made to the setup was that we took off the tape from the deck and pylon and we taped the deck on to our base.

Figure 3: Final setup; allows for bridge oscillations 

From the picture above, the deck is taped directly to the base and we believed that since the tape is on either end of the deck, it would allow the center of the deck to oscillate. Again we began running the wind tunnel up to 11m/s and everything was alright, so we gradually went up to 14m/s and that’s when we saw the bridge oscillate. Once the bridge began to oscillate, we increased the RPM by 10 and took videos of the bridge oscillation, vortex shedding, and cable oscillations.

Figure 4: Video of our bridge flow viz at ~16m/s

The video above shows our bridge at ~16m/s, which was close to the point of failure. The first thing I noticed was the airflow above and below the deck of the bridge. There was definitely vortex shedding from the top of the deck but there were also some at the bottom of the bridge. When we decided to calculate the vortex shedding frequency, it was a bit difficult to get an exact answer because from the video the vortex shedding from the top and bottom of the bridge kind of converge together after the have separated. The next thing that was interesting was the bridge oscillation, which was what we were afraid we weren’t going to see. The oscillation frequency of our bridge was very high compared to the Tacoma Narrows bridge which only oscillated 1/3 of a Hz compared to 33 Hz for our bridge. One of the reasons there is such a big difference is because of the weight and strength of the deck. Initially, we wanted to match the Reynolds number of the Tacoma Narrows bridge with our cable stayed bridge, however we found that it would not be possible to do so because of the materials we used. The deck and the string would not be able to hold at a Reynolds of 1.49e7.  

Figure 5: Experimental data from our bridge and Tacoma Narrows bridge

As you can see above, we settled on comparing the vortex shedding frequency, deck oscillation frequency, cable oscillation frequency, and Strouhal number.  These calculations showed us the relations between the vortex shedding frequency and deck oscillation frequency at different speeds. We then related this back to the calculations from the Tacoma Narrows bridge. From the data table above, the vortex shedding and deck oscillation frequency are relatively 3-4 Hz a part from each other at each speed, with the deck oscillation frequency being larger than the vortex shedding. However, in the Tacoma Narrows bridge you can see that the vortex shedding frequency is larger than the deck oscillation frequency. We believe this is due to the weight and strength of the deck. We expected the bridge oscillation frequency to be close to the vortex shedding oscillation. frequency because of the weight and strength of the deck (strength of the bridge deck needs to be much stronger for real life), therefore the vortex shedding had a much more profound effect on the bridge oscillation frequency. For our bridge the deck was made out of cardboard with some horizontal beams at the bottom of it to provide some support, so the overall strength of the deck is very minimal compared to the Tacoma Narrows bridge. In real life, the Tacoma Narrows bridge is much stronger and weights a lot more, which allows it to create a lower deck oscillation frequency than the vortex shedding frequency.

In conclusion, we found that flow separations and oscillatory behaviors are extremely Reynolds-number dependent, which we could not match using the wind tunnel. If we did, we would need a wind tunnel speed of 690 m/s! Using either CFD modeling or a variable density wind tunnel would have been more accurate.  

Figure 6: Our bridge failure; although it looks bad our bridge did not suffer any damages. The tape closest to the video could not hold, therefore the bridge flipped upwards. 

[aero307collection]

Week 4: Building our bridge for the flow viz challenge

During the fourth week of our lab my flow viz group and I began researching the Tacoma Narrows bridge to provide a comparison between the bridge we will be building the Tacoma Narrows bridge. We wanted to compare the properties of the Tacoma Narrows bridge such as its vibrations and structure to see the difference in our results when we test our bridge in the wind tunnel. In addition, I created a Solidworks model of our bridge so that we can refer to it and build the bridge to its exact dimensions. The model really helped us figure out what dimensions to use because our first design did not look the way we wanted to, so I redid the dimension and the we got a good looking design that was similar to the Tacoma Narrows bridge. The model of the bridge can be seen below. After the first half of lab, we went to the Aero Hanger machine shop to start working on our bridge. At the machine shop, I was in charge of measuring the dimensions of the wood while Kenny cut the wood on a band saw. After the wood was all cut, we measured out cardboard and began to cut the platform of the bridge.

On the second day of the week, our group went to the Mustang ’60 machine shop to finish building our bridge. We continue to cut out pieces of cardboard for our bridge such as cardboard washers and horizontal beams for the platform. After that, we measured and spaced out even holes on our platform, and the bridge arch. Then, I started to glue everything together. We decided to use a hot glue gun because we thought it would be the best option and the materials for it isn’t expensive. I glued the wood parts together to make two arches for the bridge and I also glued the entire platform together including the horizontal beams under it. This definitely took the most time because we had to wait for the glue to try for every component, even though hot glue dry’s very fast. Furthermore, we had to add more glue in some spots that didn’t get enough glue. Afterwards, Kenny and I started to string the whole bridge together. The process, was a bit challenging because we had to string the bridge from one of the platform, though the first arch, then through the middle of the platform, then though the second arch, and then finally the other side of the platform. In addition, we wanted to make the tension pretty tight so that the bridge doesn’t move too much, which was even harder because it is really hard to keep tension on a piece of cardboard. The process of building and stringing the bridge together took about three and a half hours that day, but we finished the bridge so we will be testing the bridge on Monday of next week. The only left we have to do is attach a long piece of wood under the bridge so that it has a base to stand on when we put the bridge into the wind tunnel. Since we will be running the wind tunnel had low speeds the bridge should be able to hold and not fly anywhere due to the weight of the wood from the base and the arches. I’m really happy with the design and overall look of the bridge. It looks like a well made bridge so I hope that the flow viz from it will provide us with interesting pictures and videos.

Figure 1: SoildWorks model of the bridge without the the string 

Figure 2: Gluing the arches of the bridge together

Figure 3: Cutting out the cardboard platform and gluing horizontal beams on it 

Figure 4: Alining the bridge and measuring the platform to provide equal spacing

Figure 5: Stringing the bridge and making sure to keep tension on the strings

Figure 6: Final bridge product 

[aero307collection]

Week 3: Analyzing wake test and prep for flow viz challenge

During the third week of our lab we continued working on our memo for the first lab. After finishing our wake test, we added the data and other information into our lab report. Since we divided our lab memo into two different sections, one for the first test and one for the wake test, we added the methodology, data, results, and errors for the wake test in our memo. The panel code that Chase and I created (see below) was used to calculate the coefficient of lift (Cl) and pressure drag (Cd) for both our Reynolds number as well as each angle of attack. From our calculations, we concluded that as the angle of attack increased the Cl and Cd would also increase, until stall though. At stall, the Cl would decrease for both Reynolds number, which makes sense. In the plot below, it can be seen that the lift data for our run is skewed downwards for the post stall behavior at a low Reynolds due to the peak suction occurring at an area which did not have a pressure port. Due to this, we were unable to calculate around 30 percent of our lift after stall. This only occurred for the run which our angle of attack was at 19 degrees at a low Reynolds number. All the other runs had decent data which generated accurate plots. As for the wake test, I was able to generate a very high resolution plot of our pressure data (see below). Since the plot outputted a high resolution, Robseth was able to get an accurate total drag coefficient of 0.011. Although we only were only able to take one set of data, the data was really good and I will talk about that in our memo. On the second day, Dr. Doig told us that we didn’t need to test again and that he would look for another group that had similar Reynolds and angle of attack for us so that we can use that for our memo. Jackson’s group tested similar Reynolds number and angle of attack so we will use some of their data for our lab memo.

On the second day of the week, I grouped with up a different group to prepare our second Flow Viz challenge. Initially, we wanted to use the water tunnel because it looked really interesting, however we could not find anything to test in it that would be beneficial to us. We took a long time to think of things to use because we wanted something that would have good flow visualization. In the end with the help of Dr. Doig, we decided to craft a bridge out of wood, cardboard, lots of glue and string. For first our test, we will create a bridge and put it in the wind tunnel which will be running at 5 m/s to 15 m/s. During the runs we will have smoke flowing and a laser to see the flow visualization around the bridge. In addition, we will video the runs to see if there are any vibrations or fluttering that occurs from the bridge. For our second test, we want to test the bridge at 15 m/s and 20 m/s at an angle of 0, 30, 60, and 90 degrees to see how the flow changes on the bridge at each angle. During our planning, I was responsible to sketch out our bridge design to make sure that we can build it in a short amount of time and to make sure that we can see the vibrations occur on the bridge. The design that we settle on was a suspension bridge which will be able to move and hopefully vibrate in the wind. A rough design of the bridge and its dimensions is shown below. In addition, I will be assembling and making the bridge with my group. I am really excited to see what the flow viz will look like around the bridge since it never has been done before and I hope that we can take some detailed pictures and videos to analyze.

Figure 1: Plot from our panel code

Figure 2: Wake test plot from our pressure data

Figure 3: Rough sketch of our bridge design 

[aero307collection]

Week 2: Starting memo and wake testing

During the second week of our lab we started writing our memo for the lab and we tested for a second time in the wind tunnel. On the first day of the second week, our lab group worked on starting the memo for the lab. Since there would be two parts of the lab, testing the NACA 4412 wing at different Reynolds and different angles of attack as well as testing the wake of the wing over four angles of attack, we decided to divide each part of the memo into two different sections so that we can address everything that happened within the two labs. For example, in our methodology section we divided the section up into ‘test setup’ and ‘data collection’. From those two sections we will split up the ‘test setup’ into the setup for our first test in the wind tunnel and the setup for our second test later in the week. After that we added a section on ‘calculations’, ‘test results’, ‘observations’, ‘errors’, ‘conclusion’, and an ‘appendix’. Most of the sections are filled with some information, however it is not fully completed.

After helping with setting up the memo and adding some of my XFOIL plots into the appendix for further analysis, Chase and I worked on the MatLab code to find the coefficient of pressure (Cp) and coefficient of lift (Cl) from our test last week. In order to find the Cp and Cl at each angle of attack we had to write a panel code that would display these values. Both Chase and I were in Dr. Marshall’s 306 class and one of our projects was to create a panel code so we have a little bit of experience of doing that. The code took a long time to write out since we were hard coding a lot of it and running the script and trying to make sure the outputs were accurate according to our data. We finished the majority of the code on the first day of the second week and on the second day of that week we completed the code during the first half class.

On the second day of that week Chase and I finished creating the MatLab code which would read in the data and provide us with Cp and Cl at different angles of attack. The plots we were getting at each angle of attack were similar to the ones that I created from XFOIL which made us really happy that the data we collected was close to being accurate. We were surprised to find the data to be pretty similar because all of us had not so great data in the past when we tested in the wind tunnel.

During the second half of the second day we went into the wind tunnel to do our wake test. Our plan was to test the wing at a speed of 30m/s which was at a Reynolds number of 5.21e5. The angles of attack we were going to test at was 11, 13, 15, and 17 degrees. I took charge of doing the LabView again for this test. The setup for the wake test was pretty simple and we just had to align the pitot tube with the edge of the wing. After that we moved the wing to 300mm and worked out way upwards until the wing was out of the wake. Then we moved the wing back to create the wake and collect the data points. Our first test took a bit of time because we were just learning about the process and steps. Dr. Doig told us that the next couple of tests would go a lot faster when we get use to it. Our first set of data at an angle of attack of 11 degrees went really well and we got a very high resolution plot of our data. The test afterwards at 13 degrees was when we ran into trouble. After moving the wing outside at around 370mm, we moved the wing in at around 345mm and we continued to move the wing inwards until we saw that the pressure at pressure valve 22 increased dramatically, however we did not see any change from the pressure valve all the way at 215mm. This was when we became suspicious that something was going wrong because there was not wake at all and there should’ve have been. We turned the wind tunnel to a lower speed and then Dr. Doig told use to move the wing a couple mm to see if it would move and the wing did not move at all. So the whole time we thought we moved the wing around 345mm we actually did not so we would have to delete all of the data for that test. Dr. Doig tried to to fix the machine that was used to move the wing back and forth, however the machine did not corporate in time and we only ended up with one set of data. Dr. Doig said that we can test again sometime next week so that we can get the data for our memo and to ask different groups if they have similar Reynolds number and angles of attack so that we can use their data.

Figure 1: Setup of the wind tunnel for wake testing

Figure 2: Aligning the pitot tube with the trailing edge of the NACA 4412 wing

[aero307collection]

Week 1: Prep for lab and first lab

During the first week of our lab we planned, prepped, tested in the wind tunnel. On the first day Dr. Doig provided the breakdown of the quarter and assigned us our first lab. We planned our test for the following day since we would be testing the next class period. For our test matrix we decided to examine characteristics at angles of attack around stall and in deep stall. In addition, we decided to investigate reattachment on the NACA 4412 wing by starting at a angle of attack of 19 degrees and working our way back down to 11 degrees. We will run the wind tunnel at 15 m/s and 30 m/s which corresponds to a Reynolds Number of 2.56e5 and 5.12e5 respectively. The angles of attack that we will be using are 11, 13, 15, 17, and 19 degrees.

While roles may shift, I am responsible for formulating theoretical data from XFOIL within my team. On the second day during the first half of the lab, I worked on formulating the XFOIL data and creating graphs for the different angle of attacks that we would be testing. Then we will use that data and compare it to our testing results.  Since we would be testing in the second half of class, a MatLab script was written and created to analyze the pressure at each port and then graphed to see at which angle of attack the wing would stall.

The testing in the wind tunnel was not 100 percent successful. When the wind tunnel was at 17 m/s all of the angle of attack testing went well and there was a clear sign of stall at 17 degrees. However, when the wind tunnel was at 15 m/s our team could not identify the exact stall angle. The stall angle was between 17 and 19 degrees, but when we tried to take pressure measurements between those two angles we could not find the stall angle. We will definitely talk about that in memo report and how it affects our data.

Figure 1: Theoretical data from XFOIL for 15m/s for 11, 13, 15, 17, and 19 degrees AOA

Figure 2: Theoretical data from XFOIL for 30m/s for 11, 13, 15, 17, and 19 degrees AOA

Figure 3: Monitoring the wind tunnel speeds