Race Prep Miata race-only chat.
Sponsored by:
Sponsored by:

Active rear wing test

Thread Tools
 
Search this Thread
 
Old 12-02-2014, 06:04 PM
  #181  
Senior Member
iTrader: (8)
 
cyotani's Avatar
 
Join Date: Jan 2012
Location: Azusa, CA
Posts: 1,407
Total Cats: 116
Default

Originally Posted by rharris19
The first reason is that I didn't want to have the natural state of the wing in a high drag position in case something failed and it stayed there.

Second, for us to take advantage of the wing at a higher angle in the turns, we would need a much better front aero setup than what we have or the car would have pushed.

Most importantly, I wanted to make it a passive system that didn't require the driver to think about one more thing during his 2 hour stint.

I suspected that, at the speeds we were at, we wouldn't get a whole lot out of a low drag system anyway. The air brake was our main curiosity to test functionality and reliability. It met and exceeded our expectations on both. We are still looking at a low drag setup that can be integrated fairly easily into our current if for nothing other than to play around with for grins.

In the video above, it seems as thought there is an initial delay in response from the wing, but that is because I had to switch it on to show the guys it could be turned on and off.
Wouldn't the natural state of the wing be at the normal angle of attack you set up if it were an static wing?

the course layout will have alot to do with it. It would be cool to look at the back to back comparisons on the same track and car with the air brake vs the DRS type set up.

Originally Posted by rharris19
Even if we wanted to set it up this way, we couldn't as the system to actively control it would cost enough to be prohibitive to ChumpCar's rules.
An arduino is $15 and can control the system. But if I'm understand that correctly, in the rules the added cost for an additional control module is excessive vs no added cost for activating it off a brake switch?
cyotani is offline  
Old 12-02-2014, 07:18 PM
  #182  
Elite Member
iTrader: (41)
 
rharris19's Avatar
 
Join Date: Jan 2006
Location: Seabrook, TX
Posts: 2,417
Total Cats: 20
Default

Originally Posted by cyotani
Wouldn't the natural state of the wing be at the normal angle of attack you set up if it were an static wing?
Using a stock head light motor you only have two positions out of the four :higher drag, braking, low drag, and the position you would use for a static. We chose to use it for where we would have it for a static wing and braking for the sake of simplicity for this first test.

I could easily adjust the rod down some to where it would be going between a low drag angle and higher drag for turns. The switch could be moved to the steering wheel on a button to actuate it, but this would now make it an active system that my drivers would need to pay attention to. I'm hesitant to add yet another thing to pay attention to in an endurance race.


Originally Posted by cyotani
An arduino is $15 and can control the system. But if I'm understand that correctly, in the rules the added cost for an additional control module is excessive vs no added cost for activating it off a brake switch?
You are understanding that correctly. To be fair though, to make the Arduino Board work, you will also need a GPS antenna and steering angle sensor, so the whole setup is more than just the cost of the board itself.
rharris19 is offline  
Old 12-02-2014, 08:37 PM
  #183  
Junior Member
 
Failure's Avatar
 
Join Date: Aug 2013
Posts: 180
Total Cats: -30
Default

Originally Posted by Jack Olsen
100 mph is about the slowest speed I'd be using it at. Does the news get any better at 130?
For the CoT wing, 100 MPH comes out to around Re = 800,00. From 500,000 to 1,000,000 (and beyond, presumably), nothing really changes.

That doesn't apply to your wing at all though. Different dimensions, different profile, different everything. Not even the Reynolds numbers will be the same since yours appears to have a smaller chord.

What wing do you have? If you mentioned a brand or model, I missed it.
Failure is offline  
Old 12-02-2014, 08:46 PM
  #184  
Newb
 
Jack Olsen's Avatar
 
Join Date: Jul 2014
Posts: 20
Total Cats: 3
Default

Mine is homemade.

The profile is a modified (for camber) version of NACA 63-520, with a half-inch Gurney flap.



You're way over my head on this stuff, but these are some graphs from when I was getting some help on choosing a profile.



Attached Thumbnails Active rear wing test-naca63520m-labeled1153988124.jpg   Active rear wing test-liftdata021153741218.gif   Active rear wing test-liftdragdata1153741531.gif  
Jack Olsen is offline  
Old 12-02-2014, 08:48 PM
  #185  
Junior Member
 
Failure's Avatar
 
Join Date: Aug 2013
Posts: 180
Total Cats: -30
Default

Chord and span?
Failure is offline  
Old 12-02-2014, 08:54 PM
  #186  
Newb
 
Jack Olsen's Avatar
 
Join Date: Jul 2014
Posts: 20
Total Cats: 3
Default

11" by 68-1/2"
Jack Olsen is offline  
Old 12-02-2014, 08:59 PM
  #187  
Senior Member
iTrader: (8)
 
cyotani's Avatar
 
Join Date: Jan 2012
Location: Azusa, CA
Posts: 1,407
Total Cats: 116
Default

Originally Posted by rharris19
Using a stock head light motor you only have two positions out of the four :higher drag, braking, low drag, and the position you would use for a static. We chose to use it for where we would have it for a static wing and braking for the sake of simplicity for this first test.

I could easily adjust the rod down some to where it would be going between a low drag angle and higher drag for turns. The switch could be moved to the steering wheel on a button to actuate it, but this would now make it an active system that my drivers would need to pay attention to. I'm hesitant to add yet another thing to pay attention to in an endurance race.




You are understanding that correctly. To be fair though, to make the Arduino Board work, you will also need a GPS antenna and steering angle sensor, so the whole setup is more than just the cost of the board itself.
Got it. Thanks for the info and great work getting a functional active wing tested and proven in an endurance event! I hope to tackle an active spoiler project some time down the road for my car.
cyotani is offline  
Old 12-02-2014, 09:20 PM
  #188  
Junior Member
 
Failure's Avatar
 
Join Date: Aug 2013
Posts: 180
Total Cats: -30
Default

According to tonight's napkin math, your wing creates 4 lbf of drag at 0* AoA and somewhere between 12 and 16 lbf of drag at 15* AoA and 120 MPH.

Edit: fucked up my units, sorry.
Failure is offline  
Old 12-02-2014, 09:24 PM
  #189  
Senior Member
iTrader: (8)
 
cyotani's Avatar
 
Join Date: Jan 2012
Location: Azusa, CA
Posts: 1,407
Total Cats: 116
Default

Originally Posted by Failure
According to tonight's napkin math, your wing creates 18 lbf of drag at 0* AoA and somewhere between 55 and 73 lbf of drag at 15* AoA and 120 MPH.

someone can overlay that profile view in solidworks create the airfoil and run a CFD for a quick comparison.

Does your equation take into account the gurney flap?
cyotani is offline  
Old 12-02-2014, 09:27 PM
  #190  
Junior Member
 
Failure's Avatar
 
Join Date: Aug 2013
Posts: 180
Total Cats: -30
Default

Originally Posted by cyotani
Does your equation take into account the gurney flap?
Probably not, that's just using the numbers from the graph he posted.
Failure is offline  
Old 12-02-2014, 09:27 PM
  #191  
Newb
 
Jack Olsen's Avatar
 
Join Date: Jul 2014
Posts: 20
Total Cats: 3
Default

Originally Posted by Failure
According to tonight's napkin math, your wing creates 18 lbf of drag at 0* AoA and somewhere between 55 and 73 lbf of drag at 15* AoA and 120 MPH.
So... worth it, or wasted effort?

And while I've got a smart guy's ear, what's the correct way to measure wing angle with an airfoil that's been modified for increased camber. Is it still a line from foremost to aftmost points, even if that seems at odds with the bulk of the airfoil?
Jack Olsen is offline  
Old 12-02-2014, 09:44 PM
  #192  
Junior Member
 
Failure's Avatar
 
Join Date: Aug 2013
Posts: 180
Total Cats: -30
Default

Originally Posted by Jack Olsen
So... worth it, or wasted effort?

And while I've got a smart guy's ear, what's the correct way to measure wing angle with an airfoil that's been modified for increased camber. Is it still a line from foremost to aftmost points, even if that seems at odds with the bulk of the airfoil?
Woah, I'm no smart guy. I just googled some equations and plugged in some numbers. All I know is the number I got out at the end. I can't tell you whether it's worth it or not, but a couple of pounds force doesn't really seem like that much to me. But then, neither do tenths of a second so it could still make all the difference at the track. I really don't know.

As for AoA, that's measured along the chord line straight from the leading edge to the trailing edge.

By the way, I'm not sure if you saw my edit earlier, but those original numbers were in newtons, not pounds force.
Failure is offline  
Old 12-02-2014, 10:07 PM
  #193  
Newb
 
Jack Olsen's Avatar
 
Join Date: Jul 2014
Posts: 20
Total Cats: 3
Default

In the land of the blind, the one-eyed man is at the top of the podium, even if he's Googling it.

I'm working on tenths, at this point. I've lined my wheel-wells, revised my splitter, covered more of the bottom of the car, put in this motorized wing-control contraption -- but took away enough weight to offset the new stuff. Hopefully, I'll at least be in the position of it not being worse...

I'd say I'll tell you how it works tomorrow, since I'm headed out to the track. But the forecast is NOT for a dry track. So we'll see.
Jack Olsen is offline  
Old 12-02-2014, 11:08 PM
  #194  
Senior Member
iTrader: (8)
 
cyotani's Avatar
 
Join Date: Jan 2012
Location: Azusa, CA
Posts: 1,407
Total Cats: 116
Default

I gave the CFD thing a whirl. Here's the results. The CFD was run at 120 mph with the airfoil you posted in the pic assuming the orientation of the AOA in the pic was correct.

The results: 320 lb downforce and 71 lb drag @ 120 mph (somone with experience in this field please provide a sanity check on these results).


overlayed a spline and gurney flap per your picture and scaled to your measurements.


threw in some randomly shaped end plates.


Maxed out my CPU for an hour to solve (course mesh size too)




What you really care about (71 lb drag, 320 lb downforce)



Pressure contour map at mid plane of wing


velocity contour map at mid plane of wing


Velocity flow tragectory (top)


Velocity flow tragectory (bottom)
Attached Thumbnails Active rear wing test-capture2_zps8a307fc1.png   Active rear wing test-capture3_zps115200ff.png   Active rear wing test-results_zps189c032c.png   Active rear wing test-pressure_zpsa0de8c2e.png   Active rear wing test-velocity_zps48be2c83.png  

Active rear wing test-flow1_zps6573255b.png   Active rear wing test-flow2_zpsa97fedef.png   Active rear wing test-capture_zpse77eb202.png  
cyotani is offline  
Old 12-02-2014, 11:19 PM
  #195  
Elite Member
iTrader: (1)
 
Leafy's Avatar
 
Join Date: Jun 2012
Location: NH
Posts: 9,479
Total Cats: 104
Default

For something like this I just do a 2d solve for an area thats is a convenient multiple of with wing length and multiply but the length of the wing and assume its going to be a little big higher lift and lower drag than if I spent hours waiting for a 3d solve. For my wing to get accuracy in the slot I've got to be damn fine mesh, which means 10 minutes to 2d solve or go to bed for a 3d solution.
Leafy is offline  
Old 12-03-2014, 12:05 AM
  #196  
Junior Member
 
Failure's Avatar
 
Join Date: Aug 2013
Posts: 180
Total Cats: -30
Default

Originally Posted by cyotani
What you really care about (71 lb drag, 320 lb downforce)
Well damn, now I wish I hadn't corrected myself earlier. I could have claimed to be able to do CFD calculations with nothing but a pen, paper, and an XFOIL analysis.

On topic, can you do this again at 0* AoA and post results for comparison?
Failure is offline  
Old 12-03-2014, 12:15 AM
  #197  
Senior Member
iTrader: (8)
 
cyotani's Avatar
 
Join Date: Jan 2012
Location: Azusa, CA
Posts: 1,407
Total Cats: 116
Default

Originally Posted by Failure
Well damn, now I wish I hadn't corrected myself earlier. I could have claimed to be able to do CFD calculations with nothing but a pen, paper, and an XFOIL analysis. On topic, can you do this again at 0* AoA and post results for comparison?
CFD is a useful tool but it's absolute accuracy is questionable. If you set the CFD parameters wrong the results are useless. If you set them right within 20-30% accuracy would be good results. Your math method might be closer to the actual solution. However if it leaves out the gurney flap I would question it since that's a major part of the airfoil design.



Originally Posted by Leafy
For something like this I just do a 2d solve for an area thats is a convenient multiple of with wing length and multiply but the length of the wing and assume its going to be a little big higher lift and lower drag than if I spent hours waiting for a 3d solve. For my wing to get accuracy in the slot I've got to be damn fine mesh, which means 10 minutes to 2d solve or go to bed for a 3d solution.
Ya, that would have solved quicker with a finer mesh and is probably the better way to go. I was just playing around with the program some and wanted to get some pretty 3D pictures and a ballpark figure just to test the parameters.
cyotani is offline  
Old 12-03-2014, 12:43 AM
  #198  
Junior Member
 
Failure's Avatar
 
Join Date: Aug 2013
Posts: 180
Total Cats: -30
Default

Originally Posted by cyotani
Your math method might be closer to the actual solution. However if it leaves out the gurney flap I would question it since that's a major part of the airfoil design.
Even without the issue of the gurney flap, my method doesn't consider the effects of what's going on at the wing tips. It doesn't seem like too bad of an issue for DIY aircraft hobbyists, but it is for us since we're so limited on span and have to make up that surface area with more chord. Lower aspect ratios mean that wingtip vortices affect a greater percentage of the wing, so endplate design has a huge effect here and that's an area I've found almost no reading material on. I don't even know if the effect of endplates varies with AOA or not.

For the time being, I'm just going to consider the whole matter of endplates black magic and leave it to you and your CFD wizardry.
Failure is offline  
Old 12-03-2014, 07:36 AM
  #199  
Elite Member
iTrader: (1)
 
Leafy's Avatar
 
Join Date: Jun 2012
Location: NH
Posts: 9,479
Total Cats: 104
Default

Originally Posted by Failure
Even without the issue of the gurney flap, my method doesn't consider the effects of what's going on at the wing tips. It doesn't seem like too bad of an issue for DIY aircraft hobbyists, but it is for us since we're so limited on span and have to make up that surface area with more chord. Lower aspect ratios mean that wingtip vortices affect a greater percentage of the wing, so endplate design has a huge effect here and that's an area I've found almost no reading material on. I don't even know if the effect of endplates varies with AOA or not.

For the time being, I'm just going to consider the whole matter of endplates black magic and leave it to you and your CFD wizardry.
You can still estimate what happens at the wing tips with your analysis. We have the math to do it on on air plane wings with no end plate, and you can assume that at best a perfect endplate is going to be as good as an infinite wing, do the math for both and take a stab at some percentage between the two answers, 60-70% performance for tip performance is a pretty safe bet on non-cfd'ed end plates. My current endplates were just eyeballed because my old computer couldnt handle running the sim when I designed the wing so it was never done in 3d. I've only managed to find a few percentage improvement by changing shape with the same max area.
Leafy is offline  
Old 12-03-2014, 08:18 AM
  #200  
Boost Czar
iTrader: (62)
 
Braineack's Avatar
 
Join Date: May 2005
Location: Chantilly, VA
Posts: 79,494
Total Cats: 4,080
Default

Originally Posted by rharris19
In the video above, it seems as thought there is an initial delay in response from the wing, but that is because I had to switch it on to show the guys it could be turned on and off.
this. i added the cool rock tunes to drown out my silly voice yelling at you.
Braineack is offline  


Quick Reply: Active rear wing test



All times are GMT -4. The time now is 12:20 PM.