Make a functional wing with free plywood?

In reply to Matt :

that's a Selig 1223, right?

if so, then the main benefit of it... is that it's drag is an order of magnitude less than the foils we're discussing now.  Which isn't a big deal at 60mph... but is definitely at 120mph+

sleepyhead said:

the red one... although getting to 2.3 has more to do with the thickness change, iirc

the other two plots are the "stock" cambered plate 5.5% thickness 16% and 14% camber sections with non-pointed trailing edges

by making the leading edge more rounded the flow gets accelerated around the front and onto the top "softer"... so it can 'hang on' up to higher angles... but also it won't generate a "spike"... which the others are doing.  it's important to keep in mind that the "stagnation point" on the wing will move backwards along the "lower surface" the higher you pitch up, so the flow has to make it around the leading edge.

1.8 to 2.3 is worth ~50#s on the 8ftx16" wing, at 60mph (~225lb vs ~175lb)... subject to verification

the more I've been playing around with this today, the more I'm realizing I'm wrong about this.

xfoil's changing the whole nose when I'm using it... as a byproduct of scaling off the camber line, which is "softening" the upper surface contour compared to the "Cambered Plate" coordinates.

alfadriver said:

sleepyhead said:

In reply to alfadriver :

xfoil doesn't do multi-elements

I've read some about people pulling single element data out of xfoil for each element and using it to create their own multi-element predictions... but I haven't tripped that far down the rabbit hole yet.  Drela, who wrote XFOIL, apparently wrote a multi-element code... but the license was 5? figures?  Although, that was back at the turn of the century.

but, if stafford knows of some codes that aren't as exclusive... I'm all ears

So I totally understand the rabbit hole, so this is shallow. Does the model predict where the flow would separate on the back?  If I get my very basic aero understanding right, that would be the area of the split, right?

xfoil will calculate where on the upper and lower surface the flow is transitioning from laminar to turbulent... which is not "where the flow separates from the foil".  that's not something xfoil really calculates... but you can get an idea of when it will happen by doing an AoA sweep, and then plotting the pressure coefficient graph at any angle of attack greater than where the Cl is less than the maximum.  iirc

In reply to stafford1500 :

Hey! Do it at the Challenge! Knock together a 5 page .ppt and get somebody to bring a projector. 

"Aero for Morons, or, How I Learned to stop Worrying and Love the Wing"

In reply to sleepyhead:

not a Selig 1223.

it was just called "MSHD" and apparently designed specifically for motorsports. so it looked like a good, tail heavy design for the short track i run on.

Alright, I took my time getting this post ready because it is my 1000th post and I wanted to make sure my DORK indoctrination was SPECTACULAR.
I spent a little time putting some examples together so I hope you all enjoy.

Robbie:
I am glad you started this thread and I am getting a kick out of how many people really want to dig in and get started on some projects. Your load testing plan with the cambered plate will be quite a sight and I suspect it will surprise you with the amount of force generated.

SleepyHead:
JavaFoil is my weapon of choice for 2D analysis in the share/free-ware world. X-foil was my go to but with the user-interface, it is much more of a chore to use. JavaFoil gives quite a few options for development including: Ground effect, multiple sections, easier adjustment of the profile, and most importantly in the current world SCRIPTING to make it all semi-automatic.
The first approximation for the load testing Robbie proposed could be made better, but first approximations and all that. The better method would be a vertical post Y'd at the wing to react the pitching moment AND a longitudinal strut to take the drag load. The drag strut could be instrumented as well to get real drag.

Alfa Driver:
Seperation is calculated for all of the 2D codes, but they are generally 'simplified' and don't really handle separated flow well. The other issue is that all of the 2D solvers are Steady-State. The real world works in Non-Steady-State (or transient) flow. It can be done, but as a first step it is overkill.

Matt:
That looks a lot like the Selig 1223 profile, which I am pretty happy with for low speed performance. Thin edges will kick the average home builders butt when it comes to construction.

CrustyRedExpress:
I think that has been generally suggested as a pool-side discussion group, bring beer. Love the name: may have to steal it...

Everybody else:
The pictures below show comparisons from freestream (out away from other influences) and in ground effect (near the ground - 1 chord length in this case). There are three set of images: the cambered plate that Robbie asked about in the first several pages of this thread, a typical single element wing (GOE523, which is a fairly old design), and a two-element wing (using GOE 523 twice for simplicity). The double element wing is not optimized, just thrown together to give an example. The images show the pressure fields around the wings. Red is static pressure which occurs as the air is slowed toward zero speed, Blue is low pressure. Note that the ground effect cases show the low pressure reaching out toward the ground which means the wing is trying to pull the ground up or more relative to our discussions making more downforce. The stream lines do not pass thru the ground and effectively get squeezed between the wing and ground. The stream lines give an idea of the direction of the flow around the wings with a significant downflow ahead of the wings and an upwash behind the wings. The stream lines do not pass thru the ground and effective get squeezed between the wing and ground. This is because the wings are running in low speed air and the pressure response is moving at the speed of sound, so it can translate ahead of the wing. We get this effect when a big truck drives by us and we can feel the pressure wave before the truck actually gets to our rear bumper.
Also I have plotted/scaled the Lift(Downforce) Vs Angle of attack for the configurations on similar scales so that you can see the relative effect of increased development/refinement of a wing section. Note that while the simple wing Robbie has proposed building is not the best, it does generate  downforce, just not as much as more refined designs. The take away here is SOME IS BETTER THAN NONE!

I suspect this thread will get more of you out there building aero bits and that is a good thing!
Later, Steve

In reply to sleepyhead :

If only there was a magazine for which I had designed a set of wings for a car or something similar I could document the process involved???

Paging JG...

stafford1500,

Would you like to be a design judge at FSAE Lincoln this year? It's June 20-24.

PM me if you would.

chaparral said:

stafford1500,

Would you like to be a design judge at FSAE Lincoln this year? It's June 20-24.

PM me if you would.

I already have too many obligations for that week. Thanks for the offer though. I do work with a few former FSAE and current FSAE judges. 

In reply to Robbie :

Glad you like it. Took longer to put together than I planned, but I had to kick my daughter off the computer to get the computations and images generated.

also, if you want an idea of the "3Dimensional" aspects of flow into a rear wing... you can look through this play list of an (iirc) multi-cambered APR wing mounted on the back of a FactoryFive Daytona Coupe:
https://www.youtube.com/playlist?list=PLryZziU3VoT-kdMxFkz8JU_slfAGLy78D

as you'll see... the Daytona Coupe bodywork is highly conducive to "spanwise inflow"

as it happens, the owners decided to ditch this wing for a "constant section s1223" wing with a variable height spoiler to shape the flow into the wing and get the correct angle in to the wing that way.

this is cool. im planning on doing some rudimentary data collection this year. tufting  and load data. clearly i know very little on the subject, but i have a load cell, tape and string... the load cell is 2000lbs rating and it fits in between the hatch and the body...not perfect but it will be consistent. 

so... re: MSHD & s1223

the reason they look similar is because the Graduate Advisor for the student that designed it was a Masters/PhD student under Selig... and I think was around and helped with the design on the s1223.  If not, he definitely learned the background theory used to develop that series of airfoils.  Dr. Gopalarathnam is also known some in the EAA world for developing the new KR-2S airfoils whose development was detailed/shepherded by Mark Langford.

In reply to Matt :

As noted in an earlier post concerning Robbie's proposed test setup, you are going to get some drag impact in your load cell readings. If you consider that to be part of the downforce 'Leverage' then you should be OK. Your load cell setup will give consistent readings even if they do not show you the 'real' downforce you rear wing generates. Getting the drag number for a wing as mounted would require an additional axis of measurement and then some basic math to figure out the torque effect of that drag on the downforce load cell. It is all doable with a home built setup and even possible to break the measurement up into two segments. Test with the load cell in the trunk, then stop and move the load cell to a cable that restraints the drag force and rerun your test. with those two bits of data you can get a true downforce and drag impact for your setup. There are plenty of ways to break out all the data, but they only require some basic math and fixturing.

In reply to sleepyhead :

that's excellent background info - thanks! 

To speak briefly to its effectiveness, i was running a 7" high deck lid spoiler previously and experiencing significant high speed rear wheel hopping under braking (above 100mph), chasing this problem for a couple years. When i took the spoiler off and ran the wing, the problem went away. i had the wing at 2.75 deg AoA - what a difference.

Stafford1500: i had considered the data i would be collecting to be a little dirty. But now that you have identified the specifics of how to collect it, i might set it up for a couple tests, thanks!! 

can't wait for the season to get underway!

Awesome post, thank you. We once played with ground effect on a car wing, trying both a high (trying to get clean air) and a low mount (60% of chord off the deck, if memory serves). The drivers reported the car felt better with the low wing, but knowing what I know now it may have just been the lower drag 

I like that stringline tuft test. I may have to try that. I have some vortex generators (little tiny tabs from airplanes, not the APC type) that I was also going to stick to half of the roof and see if that makes a difference. Aero is so cool.

Dabs of oil or paint can show flow lines as well if fufting isn't practical or is inconclusive.

Kieth: My big post at the top of this page wasn't a separate thread, but may develop into one. If that darn work stuff would cut me a little slack or I could find a few extra hours in the day it would be easier. If the Challenge Discussion idea keeps being brought up it may turn into a live version in October.

Keith Tanner said:

Awesome post, thank you. We once played with ground effect on a car wing, trying both a high (trying to get clean air) and a low mount (60% of chord off the deck, if memory serves). The drivers reported the car felt better with the low wing, but knowing what I know now it may have just been the lower drag 

I like that stringline tuft test. I may have to try that. I have some vortex generators (little tiny tabs from airplanes, not the APC type) that I was also going to stick to half of the roof and see if that makes a difference. Aero is so cool.

Look at this image again....

now, squint really hard.  I mean really hard.  No, I'm serious... really hard.

that first black shape is a really nicely shaped wing... or perhaps a really badly shaped car.  and, of course, if it were a car, it'd be a lot closer to the ground... which means it can be a really inefficiently shaped wing and still make downforce...

well, and sometimes they make lift...

so, moving the wing down is similar to making it a "second element" in a multi-element wing... thus it might have actually made more "vehicle system downforce", since it would be doing the same kind of things as the second element is doing in stafford's image.  This effect is more likely if the car in question had a little bit of under-floor work.  also, bringing the wing down could have effectively raised the angle of attack of the wing due to the downflow from the cabin, thereby making a higher local Cl.

so, your drivers could have been right.

which reminds me of my aircraft design professor's response to any of our questions of "should be do this?"  A:  "Well, sure, you can do that.  Just make sure to tell me why you decided to in your report."

Just for Laff's, In the last graph you suggest the Leading Airfoil could be a Vehicle and if true underbody panels could really increase Downforce , But if it were really a Car/Vehicle wouldn't/ Shouldn't that leading Shape  be turned over to More represnt Airflow over the Car and the trailing airfoil would be doing the work, I understand your Graph is a Double wing airfoil, I am just changing the first Wing to the Vehicle

GTXVette said:

Just for Laff's, In the last graph you suggest the Leading Airfoil could be a Vehicle and if true underbody panels could really increase Downforce , But if it were really a Car/Vehicle wouldn't/ Shouldn't that leading Shape  be turned over to More represnt Airflow over the Car and the trailing airfoil would be doing the work, I understand your Graph is a Double wing airfoil, I am just changing the first Wing to the Vehicle

He was suggesting the first wing element could be replaced with a car shaped part and the effect of the second element would be the wing at the back of said car. He jokingly suggested squinting to get the first element to resemble a car shape. The software I used to generate those plots will allow you to make the front shape more car-like, but that was not really the point of the images I posted.

yeah, I suppose I over did it there

but, people tend to get hung up about rear wings being "wings"... when at a certain closeness to the vehicle body they turn more into "flaps"... imho.  having that realization was important to my own conceptual thinking... so I tend to emphasize it.

That was the effect we were aiming for, definitely. It was the reason for the low mount, to use the trunk lid for ground effect. We’d also done some tuft testing and added some vents in the rear window to try to prevent separation and get some reasonable flow  

But it could also have been a drag reduction :) We didn’t get the chance to test it too much, which was the downfall of the whole program around that car. I haven’t tried to duplicate it since. 

Riddle me this.......So I have my airflow working for me and my rear wing is in Ground effect and placed Very Rearward and up or down ( I THINK HIGH WILL BE WORSE)would not say 150# Of Downforce also Lighten the front  ? Having the rear wing Behind the CL of the rear wheels,Coming off the rear deck, Ya Know a Fulcrum Effect.

 the Wing I want for the Challenge Vett Will be mounted on or Through the 'Halo' Bar (Topless Car) and Struts over the rear wheels, will be the Size if an Outlaw Sprint car, Without so Much Curve Aka Flatter, Looking For Sprint car Down force with out so much Drag,

In This Case A Giant Deflecter not really a Wing 'Cause It won't Fly.!