Aero 307 - Experimental Aerodynamics

- June 6th

Today was our last day of 307.  All of the groups went through their final presentations marking the ending of the class. 

To sum up my experiences in Aero 307 I will answer the questions that Dr. Doig suggested we reflect on.

a) If you were to pick one thing that you feel like you understand pretty well about aerodynamics now that you didn't at the start of 307, what is it and why?

One thing that I felt like I did not understand well going into 307 was the importance of the Strouhal number.  I remember in Aero 302 we had a lab or two with Strouhal number analysis and I remember being completely lost in the physical meaning behind it (as well as its importance).  After completing the baseball aerodynamics (FlowViz II) experiment I definitely felt like I got a better understanding of the subject.

b) If you were to pick one thing about aerodynamics that still confuses you, what is it and why?

This is a tough one.  There is are certainly topics still in the field of aerodynamics that I feel with uncomfortable with.  If I had to choose one I would probably relate to hypersonic aerodynamics, which isn’t exactly a part of the 307 coursework and mainly deals with gas dynamics/heat transfer issues.  To be frank, I think I just need to get more comfortable with articulating aerodynamic concepts.  I know I sort of dodge this question but unless something comes to me later that’s my answer.  Okay, actually I could brush up on wake analysis...

c) What was your 307 highlight?

Honestly I think my highlight of 307 was the interactions I had with my classmates and T.A.  I still don’t know everyone in the degree and this class definitely helped me become more antiquated with others.  In general, I found myself looking forward to hanging out and running experiments with the other 307 lads.

d) What was your 307 lowlight?

My lowlight was probably not doing as well as I would have liked on the first lab report.  In hindsight, there were several things that could have been improved upon and I definitely received valuable feedback.  Writing skills are definitely going to be important for the remainder of school and in industry and I certainly should strive to become a better conveyor of information.  I just remember being a little disappointed with myself, but I think in a way it was a “positive” lowlight because I feel like at least I learned something from it. 

e) For many of you, this will be the last time you really engage with aerodynamics, since you prefer structures or controls or design or just anything else... for others, this course will have been a springboard to many future aerodynamic adventures. What do you think the future holds for you in aerospace engineering?

Unfortunately I’m still not 100% sure what sub-discipline I would like to find myself in the future.  For my 4th year I think I would like to take Flight Test, Advanced Control Theory, and Missile Design.  From this, it appears like I am still getting a taste of all the different fields flight performance/aerodynamics from flight test, controls from control theory, and structures/aerodynamics from missile design. 

I still find it hard to believe the third year of my undergrad career is nearly over and in a few months I will be starting my last year of undergrad.  I harbor conflicted feelings of both stress and excitement for the future.  The stress comes from figuring out what exactly I want to get into after graduation and the excitement, ironically, comes from the different possibilities and paths that can be taken after undergrad.  But for now, I will just focus on doing the best I can in both my classes and life.

Lastly, I’d like to thank Dr. Doig and T.A Tynan Guerra for being great mentors in Aero 307, it was fun!

[Special thanks for the donuts, Tynan (^▽^)/ ]

- June 4 (Monday)

Today was a rather productive day for the Lost Coast FlowViz team.

Will and I arrived to the water tunnel about 45 minutes to early to see if we could get a head-start on filling the water tunnel up so we could start testing at noon.  Unfortunately, we did not know another group was going to be testing before us, so we ended up having to wait from them to finish and drain the tunnel before we could do our testing.  It didn’t really matter in the end as we started and finished our testing rather quickly.

We spent about 2 hours making sure we had all the flow visualization that we wanted from our two different fins, as well as a scale fin (that was suggested by Dr. Doig).  Luckily we did not experience any major malfunctions with equipment; although we did experiencing a slight issue with the dye coming out with a slight curve.  It was certainly easier to run through our test matrix the second time and we had plenty of time to drain the tunnel midway and fill it up for the second half of testing.

After we finished testing and cleanup, I made sure everyone had a copy of the videos and pictures from both the GoPro and DSLR footage.  The team decided that we would spend this evening doing separate tasks: Will would get started on the background information, introduction, and methodology, Yev would begin producing gifs of the static tests for comparison, and I would handle the dynamic footage and analysis.  We believe if we have a good start on our presentation we can leave tomorrow evening for the analysis and final edits. 

Below is a picture from a GIF I created showing the difference between the standard design and Lost Coast design in how quickly they return to 0 degrees from an initial deflection of 30 degrees.  (I will post the GIF after this blog).

By analyzing all the footage and exporting the subsequent data, I was able to produce several graphs that would be used in the presentation.  Below is one of these graphs:

From this graph there are two main points of analysis:

For the 30 degree test, Lost Coast’s design returned to level conditions on average 0.570 seconds faster than the standard design (or 43.6% faster)

For the 60 degree test, Lost Coast’s design returned to level conditions 0.514 seconds faster than the standard design (or 27.26% faster)

Thus a possible (and tentative) conclusion would by that the Lost Coast design turns faster than standard design.  I remember our lovely T.A, Tynan, mentioning while playing with the fins in the water that the Lost Coast designed felt like it “turned harder”, so perhaps this could help supplement those feelings.

Today our group spend the entire lab period working through our test matrix for our surfboard fin FlowViz. 

Unfortunately, we had a few issues with the dye injection equipment; however, we will be returning back on Monday to do another run through of the FlowViz test matrix.

Earlier this week, I used my housemate’s 3D printer (with permission) to print out both models of the surfboard fins that we would be working with.

The Lost Coast design is pictured behind the standard fin design.  Both models appeared to print fairly well, which I was happy with since I was worried about how it was going to handle the very fine trailing edge.  I designed the holes that would be used for fixture in the water tunnel to be 1/4″ in diameter, but they printed slightly smaller than expected.  This was not too much of an issue as we ended up just drilling out the hole to the correct size on the day of the lab.

Will and I arrived an hour before our scheduled testing time to make sure the fins would attach to the dowels and that the dowels would turn properly.  With the use of a ball bearing from a Boeing fidget spinner (thanks Boeing), Will created a fantastic mounting system with an angle indicator.

We were finally ready to begin the dynamic testing of both fins.  Our idea was that we would begin with the dynamic testing before we ended up dirtying up the water, so we could get as many runs as we wanted.  The dynamic tests seemed to work fairly well, although we did need to run the water at a higher speed to help the fins return to 0 degrees.  After running each test several times, we were ready to begin our FlowViz sequence.

Dr. Doig suggested that we should fill the water level higher and to use a new grid background with lights illuminating the background.  So we got to work setting up the additional equipment by cleaning off the new background and glass.

Unfortunately when we got to use the dye injector, a crucial component broke off, so we ended up spending about 20 minutes trying to fix it.  Even after we fixed it (with support from our favorite T.A, Tynan), the dye injector still was a bit faulty.  Pressure loss was a huge issue with our pump system, so we had to rely on using the hand-pump to keep the dye stream ejecting consistently. 

We ran through most of our static test matrix as best we could, but unfortunately there was not enough time to get it completely right.  Nonetheless, it was great practice for this upcoming Monday and I think we should be able to get even better visuals for our presentation on Wednesday.

This week was mainly focused on wrapping up the previous force balance lab report and further preparations for the Lost Coast project. 

On Monday our force balance lab worked on the lab report mainly.  We spent most of the time on analysis and overall fine-tuning of the figures for clarity.  In our analysis we found the tests to be unfortunately inconclusive.  The coefficient of lift values for the high aspect ratio wing were higher, but so was the drag.  The coefficient of drag values were an order of magnitude off from the expected values, most likely due to faulty equipment so the L/D versus alpha graphs were not particularly promising.  Initially I thought that the coefficient of drag values for the high aspect ratio wing would be lower, but the high aspect ratio wing offered more wetted-surface area that could lead to an increase in drag.  Either way, it was expected that the higher aspect ratio wing would have a better L/D curve than the low aspect ratio wing, but both appeared to be rather similar.

In other news, we have obtained and made edits to the two surfboard fin designs we are testing.  Edits include adding a plate on the top for the mounting process and ridding the Lost Coast design of internal holes.  The following images are the ‘standard’ fin and the Lost Coast design.

(The standard design)

(The Lost Coast Design)

The latter design features a less aggressive sweep design.  When talking to our TA Tynan, he told us that in his research he found that the sweep is not really for anything aerodynamically important, but rather just because it either looks cool or it is trying to match the look of an aquatic fin.

- May 16 (Wednesday)

Over the last week, we were given a task to decide which project we would like to work on for our final assignment.  The choices were between: - Automotive bluff body wind tunnel model - Surf board fin flow viz - Tesla water tunnel cavity - Re-operation and flight test of hexacopter I felt that the surf board flow viz would be the coolest project to join.  I’ve been lightly following Lost Coast Surf Tech and their efforts towards creating a better surf fin, so I was naturally interested to learn more about this topic.  Once assigned in a group with Will and and Yev (both who I’ve worked with before in this class), we began research, worked on a test plan, and asked questions to our T.A, Tynan, who is a member of Lost Coast Surf Tech. 

In our research we have discovered that fin science is currently not that well defined.  Furthermore, performance metrics for surf fins are even less defined and are instead based off of a surfer’s preference.  For example, a surfer might appreciate a board that is easier to turn sharply, while another surfer would prefer a board that turns slower.  This will make testing a little bit difficult with a lack of defined metrics to attempt to capture.  However we have decided on a test matrix that should be broad enough to be able to make some conclusions off of.

Our text matrix as follows:

As the figure suggests or test matrix is more like two separate test matrices, one for a static test and one with a static test.

The static test will be a more qualitative approach with interpreting vortex shedding patterns between two different fins.  The dynamic test was designed to extract more quantitative data from our experiment.  With the dynamic test, we can compare the two fins and see which was naturally turns faster.  Again, we can not then make conclusions towards which fin has a better performance metrics, but at least we can put numbers into our analysis.

We are going to have to construct a mount and perhaps 3D print a couple of fin models in next week, so our group will need to have talks about how we want to approach that.l

Meanwhile the force balance lab is still active and I have recently inputted our test matrix for that test into XFOIL.  Then figures below are from that XFOIL data:

- May 9th (Wednesday)

Our group for the 4412 project had our day of testing today. 

But rather than bore my reader(s?) with an overview of the experiment, what we did, are what our plans are (everything that you will be hearing in the next couple weeks anyway), I have something more interesting to talk about. 

After our 307 lab period, I went on a tour of the new residence halls that are being built at Cal Poly, yakʔitʸutʸu (which means “our community” in the Northern Chumash language).  Since I am a housing ambassador at Cal Poly I had an opportunity to have one of the first looks at a nearly completed residence hall.  While I was on the tour, I kept thinking to myself how nice these buildings were: with the floor-to-ceiling windows, fancy all gender bathrooms, and a large amount of study and communal living spaces.

We entered one of these communal living spaces and I looked up towards the large ceiling fan above us.

I suddenly realized something peculiar about this ceiling fan.

WINGLETS

Not only were these ceiling fans in an airfoil shape (I wonder how the specific airfoil geometry), they also had winglets on each of the blades! 

While looking up, I immediately and unintentionally said, “Hey, those are winglets!” and everyone around me just looked at me confused.  Then it dawned to me that I was the only aero student there and I was probably the only one who gave a damn.

While walking back out of the construction site, I was telling one of my friends about how winglets improve efficiency by reducing drag caused by tip vortices.  I then began to ponder about what the purpose for putting winglets on the blades was (besides the obvious answer of “efficiency”).  I hypothesized that perhaps the vortices shed at the end of the blades could produce a downforce on the top surface without the winglets.  This down force could then perhaps thus cause the fixed blade to bend unfavorably. 

In terms of their airfoil shape, I think it has a structural implication as well.  This ceiling fan sports rather long blades and thus might be more subject to “droop” than a smaller fan.  Assuming these blades are at a non-zero lift angle of attack, they will produce a lift force that might help counteract the bending it theoretically experiences. 

But why are these blades so long if they could have the adverse effect of structural damage to the system.  According to a NASA study:

“Similarly, for a given wing area, a high aspect ratio wing will produce less induced drag than a wing of low aspect ratio because there is less air disturbance at the tip of a longer, thinner wing. Induced drag can therefore be said to be inversely proportional to aspect ratio.”

These blades certainly have a relatively high aspect ratio and therefore will produce less induced drag, furthering the efficiency of the installed winglets.  Nifty!

- May 7th (Monday)

Today we had our presentations for our FlowViz II project.

Before the presentation my group and I met up about an half hour before to go over final details and to make 100% sure we were covering everything on the grading rubric.  A few changes including adding the physical meaning behind the Strouhal number as well as inputting our entire (initial) test matrix. 

We think the presentation went fairly well, although we might have been a little too lengthy.  This was odd because I was somewhat concerned that we didn’t have enough content.  I really enjoyed the everyone’s presentations.  I thought it was cool how everyone had very different projects, it kept it interesting!  If I had to choose a favorite presentation it would definitely be Bob and Sergio’s analysis of submarines in water.  Their methods of testing and results were fascinating and I enjoyed Bob’s secret affection for soviet submarines.

With the FlowViz project done, it is now time for our force balance studies to take place.  There has not been a lot of information (yet) about the project, but I do know my group members: Will, Sergio, and Alex. 

- May 2nd (Wednesday)

Today has been all about trying to extract quantitative data from our footage.  The footage can be seen in the two prior postings that are uploaded on this blog.  One link shows the test with the ball not spinning and the other link shows the test with the ball spinning.  We figured that we should split up the videos into separate configurations based on how we wanted to order our presentation.

One of the goals I personally had was to find the amount of dye that remained attached to the ball.  I would then compare the two different configurations and see how much of an increase in attachment the Magnus effect had with the ball spinning. 

To do this analysis I went back to my footage and grabbed a couple of still images.  Then I put both images in the program paint.net.  Using this software I was able to then measure the diameter of the ball in pixels and then compare the diameter of the ball to the point of separation.  For the no-spin configuration, I measured a diameter of 594 pixels (red line in image below) and the separation point at 386.5 pixels (purple line).

From dividing these two numbers (386.50 / 594), I calculated that the amount of dye that remained attached to the ball was 65.07%.

Next, I did the same method for the spin configuration.  The red line still represents the diameter of the ball (527 pixels) and the purple line represents the amount of flow attached (405 pixels).

With these values, I calculated that the ball experienced 76.85% attachment.  By comparison this is an overall 18.10% increase from the no-spin configuration. 

When I initially looked at the spin configuration video, it appeared that the flow stayed nearly attached for the entire time.  I noticed however that the flow towards the trailing edge of the baseball was a bit “soupy”.  While there certainly is a smaller wake left by the Magnus effect, the separation was not completely attached. 

- April 30th (Monday)

Today was not terribly exciting day in 307.

The goals that I had going into today was to shift all of the videos that we took last week, pick the best ones (so that I can later stitch them together into a concise video) and to get a good start on the presentation.  Josh and I worked together on the more foundational topics of the presentation (intro, background, methodology, etc).  Most of this information was already present in our test plan, so it was just a matter of extracting the more important bits.  Next lab period we plan on attempting on extracting quantitative analysis and then importing that into our slides.  I think the footage looks really good, but I am a little concerned on what the quality of calculations will be.  Either way, we knew from the beginning that it is more important that we have a solid analysis conceptually of what is happening in our visualization, which we have.  If we can not retrieve good working numbers of the Strouhal number we could shift our focus to what percent of the ball is seeing attached flow for the different configurations.

We do not think we should have any issue filling our 7-8 minute presentation with a quality presentation.  There is plenty of background information that we are including along with the visuals and analysis.

I unfortunately can not outdo my pictures/videos from last week, but just to give an illustration of what I was working on today, here is the lovely setup diagram I made for the presentation.

Today’s post might be a little short, but I believe I will have more interesting things to say on Wednesday.

Today our group did our testing in the water tunnel and took quite a lot of footage and pictures of the test.  I wanted to make a quick < 10 second video showcasing an observation that was seen throughout our experiment.

The first clip shows the dye over a stationary ball, while the second clip shows the flow over a ball with in a spinning configuration. 

From looking at the two clips it is obvious that the spin allows for the flow to stay attached on the ball’s surface longer than the stationary configuration.  The rotation allows for the flow to have a smaller separation wake than what can be observed in the stationary clip.  These effects can be attributed what is known as the Magnus effect.  The dye in the second clip can clearly be seen being pulled into the direction of spin, thus causing a pressure imbalance.  This pressure imbalance will result in the baseball’s trajectory being altered. 

Another interesting phenomena that can be observed in the second clip is the oscillatory vortex shedding off the back of the geometry.  This is caused by the alternating low-pressure vortices that shed off the baseball.

Now that our group has completed testing, we will begin the process of going through our media and picking our what footage we would like to use for our presentation.  Furthermore, we will be analyzing the footage and attempt to calculate quantitative data such as the Strouhal number.   

- April 23rd (Monday)

Upon arriving this the water tunnel, our group decided that rather than try to attempt testing, would would find configurations that would work best in the water tunnel.  Furthermore, we decided it would be best to use this lab period for finding wood to help mount our drill and drill holes into our 2 baseballs.

Our first step was to fill the empty water tunnel by connecting the hose from the sink inside the wind tunnel lab.  After it finished filling we turned it on straight away.

We started the pump and the water began flowing at the maximum speed of 5 inches per second.  After applying the dye inside the stream we noticed that by the time the dye reached the test section area (marked with the glass) the dye was already a turbulent mess.  Our group then moved the speed down to around ~3.8 inches per second and found that the dye stayed attached for much longer, thus would be more useful in getting good streaklines.  Even though we did not have our baseballs drilled yet, we decided to just hold a baseball inside the section and see what the ball looked like with some dye coming at it.

Another problem we had to consider was keeping the dye injector from moving around in the water.  However, with just a little masking tape we were able to fix the injector into a fixed position and Bennett was able to get the hand pump delivering a consistent amount of dye to be thrown over the baseball.

After we were satisfied with learning how to operate the water tunnel we then began the rather arduous task of draining the water back into the sink [not the drain].  While Josh made sure the draining process was running smoothly, Bennett and I went to Mustang 60 and drilled holes into two of our baseballs.  One was drilled so that the spin axes was in a two-seam condition and the other in a four-seam condition.

We are definitely glad that we decided to take today as more of a “prep day” before we actually started running through our test matrix (which is now set for this Wednesday).  If we did not run through the water tunnel before our testing time, we might have wasted precious time trying to find a better way to handle the dye injection unit and find a better speed to run at.

- April 16th (Monday)

During Monday’s lab we had our last day of testing with the 4412 wing.  Today’s goal was to capture coefficient of data pressure to show the effects it has on the wake.

We were able to run through our entire test matrix for the 15 m/s condition, unfortunately we were not able to get through our entire 30 m/s condition.  This was due to some issues capturing the wake during the 15 m/s condition. 

Before testing we did some trig calculations in order to figure out the position of the wake so we knew which points to test along the wing.  During the actual experiment however, we did miss the wake a few times so we had to go back and retest a few angle of attack configurations.  While we were disappointed that we didn’t get through our entire test matrix, we all still feel like we collected enough data to have a good write up for this section of the report.

In regards to the report, we will be working on it this Friday as we would like to get it down before we all get too distracted by our FlowViz projects.

- April 18th (Wednesday)

The purpose of today’s lab section was to come up with a test plan for the flow visualization project  In the days prior to this lab section we had some trouble thinking of what exactly we wanted to test, but fortunately we have landed on a subject that we think will be a good test for us.  Josh, Bennett, and I will be testing the effects of spin towards the aerodynamics of a baseball.  The goal section from our test plan is as follows:

Study and compare the flow around a baseball with a low spin vs a high spin.  Configurations from the axis of spin (2-seam and 4-seam) will also be compared. Separation of the flow at the top and bottom of the spin will be observed and cataloged.  Strouhal number analysis between the two testing conditions and observation of the magnus effect and the consequences it has toward the aerodynamic characteristics of a baseball in flight will be performed.  Results will be compared to the known results of balls that are pitched in this style.

We will be using the water tunnel in order to capture our flow visualization.  No one in our group has ever worked with the water tunnel, however we took the time today to check it out in the yard next to the wind tunnel.  Initially we struggled with how we wanted to configure out set up, but we have settled on using a drill to spin from the top of the water tunnel.  Furthermore sometime before testing (hopefully on Monday) we are going to need to drill press two baseballs in different spin configurations.

One thing that I found particularly interesting in my research about baseball spin is that some pitchers can put a spin on ball up to 2,500-3000 RPM. 

These are some of the graphs that I produced from the XFOIL data that I collected.  According to these plots, it appears that max L/D for a NACA4412 at 30 m/s occurs at roughly at an angle of attack around ~7 degrees.