Welcome to the North American Subaru Impreza Owners Club Monday July 23, 2018
Home Forums WikiNASIOC Products Store Modifications Upgrade Garage
NASIOC
Go Back   NASIOC > NASIOC General > Motorsports

Welcome to NASIOC - The world's largest online community for Subaru enthusiasts!
Welcome to the NASIOC.com Subaru forum.

You are currently viewing our forum as a guest, which gives you limited access to view most discussions and access our other features. By joining our community, free of charge, you will have access to post topics, communicate privately with other members (PM), respond to polls, upload content and access many other special features. Registration is free, fast and simple, so please join our community today!

If you have any problems with the registration process or your account login, please contact us.
* Registered users of the site do not see these ads. 
Reply
 
Thread Tools Display Modes
Old 05-12-2018, 04:28 PM   #1
corona672
Scooby Newbie
 
Member#: 262902
Join Date: Nov 2010
Location: Arlington, VA
Vehicle:
'04 WRX Hybrid
It's a wrap!

Default Rear Wing Design, because science

Iím making a rear wing for NASA TT4, and wanted to document for anyone coming along the same path down the road. Suggestions are very welcome.

1. Scope

This is not intended to be a journal article, but more to spark discussion and other fabrication projects from other readers. I havenít taken a fluids course in quite a few years and I donít expect to hit all the analysis perfect.

The goal for this project was to keep the wing design simple, and cheap. With existing scrap material (Al sheet), the budget is <$150 for a wing fully fabricated. The front splitter design and construction isnít included. For $150 I thought it would be a fun project to work on, even if it didnít come out well. Iíll be documenting up to section 3, Fabrication Ė Iím hoping to get in feedback before that step.

The application is for a 2004 WRX currently with oem aero, in NASA TT4. The car is reasonably quick on MOTON 2 ways and Hoosiers, and to stay in class with aero the car must be detuned ~10 average HP or gain 100lbs. My minimum weight is currently 3131 with ave HP of 269. Iím hoping for a 1-2 second gain at Summit Point, down into the 1:19ís.

2. Technical Approach

2.1. Design Constraints and Assumptions

The wing is being designed in accordance with NASA Time Trial 4 rules, which restricts overall wingspan, wing location, wing type, and endplate dimensions.

The wing must not be larger than body width, no higher than 8Ē off the roofline, no further back than 12Ē from rear of vehicle, single airfoil design, and endplates no larger than 12x12Ē. This information is what I used to set my chord and span. More rules section 6.4.2 https://nasa-assets.s3.amazonaws.com...4--3-21-18.pdf

Computational fluid dynamic software like any simulation software is great if you have your model and mesh setup correctly. Numerical solutions can be accurate, but I always assume they are not and use the software to compare how a change in the model affects the solution (did it increase L/D, by how much %, est.).

2.2. CAD


I started by taking initial measurements of the car to generate a model of the wing and the body shape. I used a plumb line and a 2x6 to get the body width and height of the uprights. I ended up with the front body width to be 69.25Ē and the rear to by 68.75Ē. For user error and tolerance stack up, I chose 68.5Ē for my total width including endplates.

I wanted an airfoil that had high lift/low drag at low Reynolds numbers, approached stall angle gently, and was a known airfoil shape in motorsports. The most promising airfoils were S1223 and E423. E423 is in red and S1223 in Green (from Airfoiltools.com)


E423 in magenta and S1223 in Yellow @ Re = 500,000


I selected E423 as it is slightly less aggressive in stall angle and doesnít taper the trailing edge as much which had me worried for fabrication. Itís also a more efficient wing and the car doesnít make much power to overcome a ton of drag.

Chord was selected by leveraging documentation from technical notes and existing products. I used an aspect ratio of ~5.6 which left me with a chord of 12Ē. This was also the max endplate size, and I didnít want my chord to stick out past my endplate

Heres the CAD from SolidWorks, the endplates and uprights are a bit outdated. The endplates can rotate about 20 degrees now and the uprights are changed to accommodate fitment into the body.






2.3. Construction Process

The wing core will be from Flyingfoam.com Ė they can hot wire a custom airfoil spec in 2#EPS for pretty cheap, but this was half my budget. The ribs are made from .25Ē aluminum and thereís two 3/8Ē aluminum tubes from Mcmaster Carr that run the span for locating the parts together and for extra support. Essentially, Iíll have three foam cores sandwiched with aluminum. Glue it together, sand, and then ~2layers of 4oz fiberglass over it all. Itís really the fiberglass sheeting that gives the support. I havenít dabbled much with fiberglass, but how hard could it be?

The uprights are made from .4375Ē aluminum - Iíd prefer these to be thinner but itís what I have. Via adjustment on the upright, the airfoil range normal to the wind direction can be adjusted over 31 degrees, in 1 degree increments. Iím chassis mounting the uprights through the trunk with Al angles and an Al plate bolted in the trunk.

Ideally, id just send the metal to be laser cut or CNCd out from a local shop, but that cost between $100-$200 for all the work I figured. Iíll try with a dremel and a Bridgeport mill myself.
* Registered users of the site do not see these ads.
corona672 is offline   Reply With Quote
Sponsored Links
* Registered users of the site do not see these ads.
Old 05-12-2018, 04:28 PM   #2
corona672
Scooby Newbie
 
Member#: 262902
Join Date: Nov 2010
Location: Arlington, VA
Vehicle:
'04 WRX Hybrid
It's a wrap!

Default

2.4. CFD
I used ANSYS for my CFD software.

I tried to model the aero to a reasonable level, but left room for improv when it’s actually built. The mesh is not of high quality and I’d like to make a new model of the WRX that’s smoother. Unfortunately, my computer like to crash occasionally so I setup the mesh and solution setup to converge faster.

Setup: I used a viscous K-epsilon model at speed of 60 and 100mph. At summit, average speed on track is 88, lowest speed through corner is 45, highest through corner is 100, top speed is ~132. Reynolds number goes between 1,150,000 and 384,000. A good average would be at 88mph which is Re = ~750,000

I first looked at the effective wind angle to the airfoil compared to the WRX on level ground. Due to the shape of a car’s rear body, the airflow at the wing already has a positive AOA . This is more apparent at the center of the car vs the outside, where the air can go around the windows some. Here are some pictures I found online showing the effective wind angle to an airfoil.



Another one: https://i.imgur.com/uc4N6z2.png

I had one more that showed angle of effective air at the end of the wing span was 3.5 degrees, but it became a corrupt file :/ Heres a screen grab: https://i.imgur.com/u2jntq8.png


So what I determined from this was there is a relative wind angle greater than zero AOA to the airfoil (X) and another angle greater than zero but less than X at the outer sections of the span (Y). I assumed X = 12 degrees and Y = 3 degrees, a delta of 9 degrees. This is why you see 3D airfoils for rear wings, to help get that delta down by varying AOA along the chord. My 2D wing will be at bit unbalanced (some aifoil at or near stall, some with less Cl, so the higher I am designing the uprights to put the wing into a uniform air distribution at max allowed height.

I ran the wing by itself at various AOA (pressure from 12, 14, 16, 18, 20, 25, 30 AOA


Velocity streamline at 10,12,14,16,18,20,25,30 AOA


Pressure on the airfoil and more velocity streamlines



I’ll add a 1% or 2% chord gurney flap, didn’t get around to running the simulation. The gurney flap should allow higher AoA which would generate a higher downforce.

Data at 60 and 100 mph (no gurney flap)








From observing the Cl plot, the wing doesn’t generate much more lift past 14 degrees (stall). Since the center portion of the wing is already seeing an effective AOA of ~12 degrees, I wont be able to run much actual AOA without inducing stall in some portions of the wing. Likely I’ll start with 10 and work may way down towards 0 AoA to dial in in.

I’ll link to these when I start fabricating and testing over the next month or so.

3. Fabrication
4. Testing

Last edited by corona672; 06-11-2018 at 07:21 PM.
corona672 is offline   Reply With Quote
Old 05-12-2018, 10:04 PM   #3
WR Blue
Scooby Newbie
 
Member#: 451536
Join Date: Jul 2016
Location: Bahamas
Vehicle:
07 WRX
WR Blue

Default

sounds good. it looks good. don't get to great an angle because u may end up with zero effct
WR Blue is offline   Reply With Quote
Old 05-14-2018, 07:45 PM   #4
Homemade WRX
Scooby Guru
 
Member#: 33782
Join Date: Mar 2003
Chapter/Region: MAIC
Location: Mooresville, NC
Vehicle:
96 3MI Racing LLC
search FIRST, then PM!!!

Default

Kind of surprised you're not doing a swan neck, especially looking at your construction method.

I'd been looking into flyingfoam.com for an unrelated project of my own.
Homemade WRX is offline   Reply With Quote
Old 05-14-2018, 10:09 PM   #5
corona672
Scooby Newbie
 
Member#: 262902
Join Date: Nov 2010
Location: Arlington, VA
Vehicle:
'04 WRX Hybrid
It's a wrap!

Default

Went back and forth between a swan neck - had it all drawn up at one point. But the uprights have to be big and I don't quite have the length for it with the Al sheets I have.

I may try again and get creative as to where I adjust the AoA so a big piece of Al is not needed - but trying to keep it simple. I'm not sure if I'll need more downforce, you can only run a 4" splitter in TT4 so the front won't generate much to counterbalance.

The latest rev. TT doesn't stand a chance, Pew pew!


Quote:
Originally Posted by Homemade WRX View Post
I'd been looking into flyingfoam.com for an unrelated project of my own.
I am very curious..
corona672 is offline   Reply With Quote
Old 05-14-2018, 11:46 PM   #6
corona672
Scooby Newbie
 
Member#: 262902
Join Date: Nov 2010
Location: Arlington, VA
Vehicle:
'04 WRX Hybrid
It's a wrap!

Default

Fancier parts = exponentially more time using dremel.

corona672 is offline   Reply With Quote
Old 05-15-2018, 12:31 PM   #7
Homemade WRX
Scooby Guru
 
Member#: 33782
Join Date: Mar 2003
Chapter/Region: MAIC
Location: Mooresville, NC
Vehicle:
96 3MI Racing LLC
search FIRST, then PM!!!

Default

Quote:
Originally Posted by corona672 View Post
I am very curious..
Can't really say too much about it for now. I've actually been trying to get quotes on some larger CNC router tables with no luck. About to just pull some funds and order one.

Long story short, this project is on a fast path for a ~100,000 ft drop in July for some data collection.

Might share some pics down the road.
Homemade WRX is offline   Reply With Quote
Old 05-15-2018, 09:57 PM   #8
corona672
Scooby Newbie
 
Member#: 262902
Join Date: Nov 2010
Location: Arlington, VA
Vehicle:
'04 WRX Hybrid
It's a wrap!

Default

Maybe get some 80/20 and make your own table? A few DIYs around. That sounds like a fun project - time to partner up with a CubeSat mission.

I ordered the core from Flying foam today but theres a lead time of 1-2 weeks and I doubt I'll be able to make Summit in June with the splitter in wing installed. Also - I can't make hyperfest this year
corona672 is offline   Reply With Quote
Old 05-16-2018, 12:23 PM   #9
Homemade WRX
Scooby Guru
 
Member#: 33782
Join Date: Mar 2003
Chapter/Region: MAIC
Location: Mooresville, NC
Vehicle:
96 3MI Racing LLC
search FIRST, then PM!!!

Default

Well you hit the key word, CubeSat. So now you've got an idea of what we're working on.

I already have my own 2'x4' Gerber water jet table...guts. The shop it came from went larger on the table.

So I've started my build plans for making it a 2'x4' and roughly 5" on the Z-axis for now. by retrofitting a motor to their manual Z. I'll have to see how the rest of the table comes together.


Have you been following Nine Lives Racing wing project? Pretty cute idea that they went to manufacturing a real foil in aluminum extrusion.
Homemade WRX is offline   Reply With Quote
Old 05-17-2018, 09:02 PM   #10
corona672
Scooby Newbie
 
Member#: 262902
Join Date: Nov 2010
Location: Arlington, VA
Vehicle:
'04 WRX Hybrid
It's a wrap!

Default

Cool stuff - finding a good stepper motor may be the trick.

I have't heard of Nine Lives before, it's interesting that they made the NP01 wing. So they fix the chord dimension and only vary the span, which means 1 die. I wonder what the setup and tooling costs were for that first run... but it must make it much less hassle to get the airfoil in at the shop all perfectly finished.
corona672 is offline   Reply With Quote
Old 05-18-2018, 10:40 AM   #11
Homemade WRX
Scooby Guru
 
Member#: 33782
Join Date: Mar 2003
Chapter/Region: MAIC
Location: Mooresville, NC
Vehicle:
96 3MI Racing LLC
search FIRST, then PM!!!

Default

Yeah, I know aluminum extrusions can vary pertty widely but I can't imagine this one was terrible.

I'm wondering if they'll start to venture into other foil options.
Homemade WRX is offline   Reply With Quote
Old 05-19-2018, 12:24 AM   #12
corona672
Scooby Newbie
 
Member#: 262902
Join Date: Nov 2010
Location: Arlington, VA
Vehicle:
'04 WRX Hybrid
It's a wrap!

Default

A die cast airfoil with varying AoA would be interesting..


First rib, needs sanding.This is going to be a big wang.





Didn't like where the swan neck adjusted AoA to fit within the metal I had, so standard bottom mount.
corona672 is offline   Reply With Quote
Old 05-19-2018, 05:32 AM   #13
mpen747
Scooby Newbie
 
Member#: 292343
Join Date: Aug 2011
Chapter/Region: SCIC
Location: temecula
Default

Quote:
Originally Posted by corona672 View Post
Maybe get some 80/20 and make your own table? A few DIYs around. That sounds like a fun project - time to partner up with a CubeSat mission.

I ordered the core from Flying foam today but theres a lead time of 1-2 weeks and I doubt I'll be able to make Summit in June with the splitter in wing installed. Also - I can't make hyperfest this year
You might wanna check your nearby universities around your area. FSAE/Design Build Fly guys might be able to help with your airfoil foam cutting.

Good luck with the project!
mpen747 is offline   Reply With Quote
Old 05-19-2018, 05:38 PM   #14
Patrick Olsen
NASIOC Supporter
 
Member#: 120
Join Date: Jul 1999
Chapter/Region: AKIC
Location: Where the Navy sends me...
Vehicle:
1997 Legacy 2.5GT
1996 Impreza coupe

Default

Quote:
Originally Posted by corona672 View Post
First rib, needs sanding.This is going to be a big wang.

https://i.imgur.com/7d1R04P.jpg
That's some serious Dremel work!

Thanks for sharing all of the data analysis. Makes me feel stupid, but interesting to read nonetheless!
Patrick Olsen is offline   Reply With Quote
Old 05-20-2018, 11:29 PM   #15
corona672
Scooby Newbie
 
Member#: 262902
Join Date: Nov 2010
Location: Arlington, VA
Vehicle:
'04 WRX Hybrid
It's a wrap!

Default

Quote:
Originally Posted by Patrick Olsen View Post
That's some serious Dremel work!

Thanks for sharing all of the data analysis. Makes me feel stupid, but interesting to read nonetheless!
Jig saw ended up being a lifesaver - just kept chewing through dremel wheels. I'm not looking forward to the uprights..

Quote:
Originally Posted by mpen747 View Post
You might wanna check your nearby universities around your area. FSAE/Design Build Fly guys might be able to help with your airfoil foam cutting.

Good luck with the project!
Foam comes cut with the airfoil made and holes wired out too! Good idea though, I'm sure they have made a few of their own cores and have a solid layup procedure done.
corona672 is offline   Reply With Quote
Old 05-21-2018, 10:01 AM   #16
Hawkeye1231
Scooby Newbie
 
Member#: 483440
Join Date: Mar 2018
Default Wow

Beautiful work
Hawkeye1231 is offline   Reply With Quote
Old 05-27-2018, 04:18 PM   #17
corona672
Scooby Newbie
 
Member#: 262902
Join Date: Nov 2010
Location: Arlington, VA
Vehicle:
'04 WRX Hybrid
It's a wrap!

Default

3. Fabrication

Not quite done with the fab but here's an update.

Both the inner ribs framed with aluminum rods.


Gluing the inner and outer ribs to the foam to keep everything in place for the fiberglass layup. Outer ribs have pemnuts for the endplates.


The wing sandwich made - the rod has some flex, might consider using a larger hollow tube to keep weight down and moment of inertia up next time. Maybe use a rectangle spar in the wing, but I'm hoping the fiberglass will stay rigid. Used LW spackling to fill in a the litte gaps, but the foam was cut perfect and everything lined up real nice.


My makeshift work bench using the trailer...


Using a mix of 1:1 scale templates or using a caliper. Makes for some not-so straight cuts using the jigsaw. I would probably not cut all this thick metal out be hand again, having these milled out would be worth the few hundred $'s.


More metal bits


Just need the layup and brackets for the chassis mount! Oh and the front splitter
corona672 is offline   Reply With Quote
Old 05-30-2018, 06:29 AM   #18
mpen747
Scooby Newbie
 
Member#: 292343
Join Date: Aug 2011
Chapter/Region: SCIC
Location: temecula
Default

Hmmm for what it's worth, I feel like you could optimize/trim the ribs more. Are you gonna mount that thing on the trunk itself?
mpen747 is offline   Reply With Quote
Old 06-08-2018, 11:18 PM   #19
corona672
Scooby Newbie
 
Member#: 262902
Join Date: Nov 2010
Location: Arlington, VA
Vehicle:
'04 WRX Hybrid
It's a wrap!

Default

Quote:
Originally Posted by mpen747 View Post
Hmmm for what it's worth, I feel like you could optimize/trim the ribs more. Are you gonna mount that thing on the trunk itself?
Definitely could have, didn't want to spend the time. I'd change a few things the next go round (rectangle spar down span for flexing, slightly wider mounts, fiberglassing technique) The uprights mount through the trunk onto the wheel well floor with a base made of plywood, 1/4" Aluminum sheet, and Al angle. I can jump up and down on the base and pull all around on the uprights - not going anywhere.

So here's how it turned out! Turns out I'm not the best at glassing, so finish isn't phenomenal but good enough.



(not shown - the smallest gurney flap ever)


With the splitter added


Summit point is this weekend - I'm concerned about excessive force on the outer span on the wing and the splitter being too low, I'll find out how it goes!
corona672 is offline   Reply With Quote
Old 06-11-2018, 10:53 AM   #20
Plastic Robot
Scooby Newbie
 
Member#: 486551
Join Date: May 2018
Default

It looks like it turned out really good. Definitely report back about how it holds up at the track.

I'm inspired to try something similar and it would be great to hear more about the fiberglass process you used.
What kind of fiberglass and how many layers did you end up using?
Did you wrap each layer of fiberglass around the leading edge of the wing?
Do any vacuum bagging or resin infusion?
Plastic Robot is offline   Reply With Quote
Old 06-11-2018, 07:18 PM   #21
corona672
Scooby Newbie
 
Member#: 262902
Join Date: Nov 2010
Location: Arlington, VA
Vehicle:
'04 WRX Hybrid
It's a wrap!

Default

For fiberglass the material was 4oz e-glass. I started by running 1 layer ~ 3" width before, after, and through the rib (you can see the little slot in the middle) down the entire span on the bottom of the foil. I did two additional layers of a full sheet by dry-layup on one side, flipping, and then dry layup over the other side so the leading edge was wrapped without a break. I didn't vacuum bag or infuse but I would suggest to do so if you have the means - it was hard to keep the fiberglass to stop from peeling away from the foam when gravity took over after flipping. If I didnt bag it I would do smaller sections at a time, cure, and then do another section to prevent the glass from pulling off the foam.

After doing this, there was still some flex in the wing. Test weight via random rotors I had around and myself to about 350 lb went good. I contemplated adding a rectangle spar with a high area moment of inertia down the span to give it some rigidity, but bought some 2" wide 5.7oz uni carbon fiber instead to run down the span top and bottom of the airfoil after the fiberglassed cured instead. I'd probably add a spar next time - the Al stingers help but still not the best. Was noticeably better after even the small amount of CF. I could probably flip the car over by pushing on the uprights enough, chassis mount is going nowhere.


4. Testing

Summit point went well - Splitter and wing held up but the last minute gurney flap came a bit unglued from the paint. I'm going to buy a slightly larger steel angle and bolt/rivit it through the trailing edge - which should stiffen it up more too. I thought something looked familiar when the debris flag came out..

Wasn't able to squeak out a PR but after looking through the data I picked up 4-6mph in the fast turns (T3, T4, T9) and was able to brake at least a half marker later at T1. Started at 10 deg AOA which was understeery and turned down to 3deg at the end of Sunday, which felt nice after loosening the front sway bar up slightly too. Looking at the data I lost myself accelerating out of the turns getting on the gas. Not to make too many excuses - but the temp out was 35 deg hotter then my PR in april & humid all weekend and I'm up to ~30 Heat Cycles on the Hoosiers. Good enough for first place anyways

More pictures to come soon!
corona672 is offline   Reply With Quote
Old 06-14-2018, 01:53 AM   #22
speedyHAM
Scooby Specialist
 
Member#: 48377
Join Date: Nov 2003
Chapter/Region: MWSOC
Location: "They eat fish soaked in lye"
Vehicle:
1996 Gutted, built
XP class Impreza L

Default

Awesome. Can't do a dual element wing in TT4? You probably don't need it as you are getting enough downforce at a fairly low AoA on the current wing. Larger endplates seems like they would be helpful as well if they were legal.

Excellent writeup and great engineering work. Please share more if so inclined.
speedyHAM is offline   Reply With Quote
Old 06-14-2018, 10:28 PM   #23
corona672
Scooby Newbie
 
Member#: 262902
Join Date: Nov 2010
Location: Arlington, VA
Vehicle:
'04 WRX Hybrid
It's a wrap!

Default

Only single elements in TT4, and the endplates are restricted to 12x12" (mine are currently 12"x11"). Maybe I'll make a few more shapes as those pieces where easy to cut out.

Photo cred to Finish Line Productions http://www.finishlineprod.net/nasa/ - I need to buy a set after a few more events.





The splitter did rub slightly coming around turns and braking - think T4 at Summit. I'm not too worried about it but I can shorten up the threaded rod/turnbuckle to raise the front of the splitter up marginally. Also - the sheet metal for the front air dam will pretty much need new paint every event. Those stupid vettes in front of me just love throwing rocks up. May use ABS here in the furture. (1/8" thick??)

Not much more to share! Anything specific I'd be happy to add info/pictures to.
corona672 is offline   Reply With Quote
Old 06-17-2018, 08:09 PM   #24
Sgt.Gator
Scooby Newbie
 
Member#: 374688
Join Date: Nov 2013
Chapter/Region: NWIC
Location: Bend Oregon
Vehicle:
2005 STI
WRB

Default

Great looking race car! Is there a build thread?
Sgt.Gator is offline   Reply With Quote
Old 06-18-2018, 10:53 AM   #25
gills
Scooby Specialist
 
Member#: 11316
Join Date: Oct 2001
Chapter/Region: Tri-State
Location: Long Island, NY
Vehicle:
1988 E30S52 DD
S14VQ35 & S13KA racecarzz

Default

Quote:
Originally Posted by corona672 View Post
...Wasn't able to squeak out a PR but after looking through the data I picked up 4-6mph in the fast turns (T3, T4, T9) and was able to brake at least a half marker later at T1...
Straight away speed leading up to T1 any different?
gills is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump

All times are GMT -4. The time now is 07:36 AM.


Powered by vBulletin® Version 3.7.0
Copyright ©2000 - 2018, Jelsoft Enterprises Ltd.
Powered by Searchlight © 2018 Axivo Inc.
Copyright ©1999 - 2017, North American Subaru Impreza Owners Club, Inc.