Poor Man's Aero: Building Your Own Wing (many pix)

[Eric-325I]

Awesome and quite inspiring! Thanks for sharing... Love your 911 by the way :-)

Eric

[Jack Olsen]

Originally posted by randywebb
How thick are the tubes?

They're .5 inch outer diameter (6063 alloy). The plastic pieces snapped over for reinforcement are PVC pipe with a slightly-less-than .5 inch inner diameter. The sliding insert tube that goes inside the fixed tubes is .345 inch outer diameter.

But the lateral rigidity comes mostly from the spar pieces, which have a T-shaped profile (5/8 inch by 1/2), with glued-in bridge sections to create an I-shaped beam that runs the length of the wing. Even on their own, the T-shaped pieces (6061 alloy) are surprisingly rigid.

[David]

Very nice, looks like you really did your homework on this project. It sure is nice that NACA makes their data readily available to us.

[Jack Olsen]

Here's how it breaks down for travel:



The black piece is the diagonal support. It's a threaded bar with an airfoil-shaped cover over it, to reduce drag.

The aluminum tube next to it is the one that fits inside the tube that runs between the braces in the center of the wing. If you look at the ends of the wing pieces, you can also see the rear tube insert (it's a little washed out in this photo), which has the hole for the pin that holds the two halves in position when they're assembled.

And here's a picture of the base pieces (they also come off for normal driving) from the side:



And from below:



The key part about the base is that there really isn't much stress placed on the fiberglass decklid. The steel braces are sandwiched down on top of the body of the car (along the edges of the engine compartment) so that the force goes down, right to the chassis steel. The T-shaped front piece is very sturdy, which controls fore-aft movement. The rear mount is just there to distribute force down to the chassis. The base pieces abrade the paint somewhat, but it's the paint deep down in the gap between the decklid and the chassis, which isn't visible when the decklid is down.

[911pcars]

"The base pieces abrade the paint somewhat, but it's the paint deep down in the gap between the decklid and the chassis, which isn't visible when the decklid is down."

Aha. A place to improve!
A piece of inner tube or any convenient squishy material glued here should prevent further erosion of paint and metal. Punch appropriate cutouts to access the bracket fasteners.

Happy to contribute to the engineering of this.

Sherwood

[Jack Olsen]

That's a good idea, Sherwood. I use adhesive-backed felt pads on the parts that could conceivably touch the exterior paint. But something dense and slightly elastic (like inner-tube material) underneath would eliminate damage down there.

[Richard Burns]

Well done Jack. My understanding of the diffuser is that it doesn't need to be as long as the one you have already tried. Look at a surfboard with 3 fins that is enough to do the job and I believe that is all that is required on a car.Perhaps your aerodynamisist could correct me if I'm wrong.
Richard.

[Jack Olsen]

A weight update: Total for the wing, uprights, caps, hardware and ducktail itself: 22 pounds.

That's less than half the weight of my old setup (45 pounds). And it's weight that was far back and high up.

The old carbon fiber wing was 9 pounds all by itself, and it was 12 inches shorter than the new 4.75 pound one.

[kach22i]

Nice work.

[Jack Olsen]

Weight testing report: I put 250 pounds (five 50-lb bags of sand) on the wing today, with some cushioning to distribute the weight more evenly over its surface. It turns out that I have to make my connecting tube more rigid (maybe steel instead of aluminum?), and distribute the load to the fiberglass decklid a little more (I started to hear the fiberglass complaining).

But nothing got damaged, and the wing itself held up fine, which isn't bad for a 4 3/4 pound collection of thin aluminum pieces.

[911pcars]

"...and distribute the load to the fiberglass decklid a little more (I started to hear the fiberglass complaining). "

Maybe keep the basic attachment location under the FG lid, but make the base longer/shape it so when the wing is compressed by air pressure, the load transfers to the chassis rain gutter directly under the mount instead of straining the limits of FG strands. You could also shape one base edge so it sits over (U-shape) the lip of the engine compartment edge. Might add some rigidity to reduce any twisting tendencies of the upright brackets as car direction goes off-axis @ speed. Or just mount the uprights in the side gutters of the more-rigid eng. compartment opening. The height of the wing should allow the engine lid to fully open. Need a lightweight CF engine lid support? :-)

FWIW, F1 pays a lot of attention to vehicle dynamics during high speed cornering, not only straight line aerodynamics. Can you measure change in attack angle or other wing/bracket deflection during high-speed cornering with the equipment you have (e.g. strain gauges)?

Sherwood

[Jack Olsen]

Yeah, that's my plan. The downward force actually does what you're describing -- it pushes down onto the chassis underneath the decklid. But the way the wing extends back means that downward pressure on the wing also pulls up on the lid -- imagine pulling straight back on the wing it wants to lift the forward edge of the decklid. It's like a lever effect.

I'm going to keep the existing mounts, but also integrate in a triangle-shape on the side that folds down below. It will run up a little higher on the rain gutter part of the chassis, but it will also distribute the force more evenly to the decklid itself.

[Zeke]

I've been making home brewed things all my life from the first skate boards to be seen, to 20" bikes with kart motors welded above the removed pedals. Then, of course, the car thing, now for some 45 years and probably 45 cars. I've never been more impressed. With the amount of time that you've spent with your tools compared to me and the fact that I'm a journeyman carpenter with $1000's in tools, it's a real accomplishment to conceive, design and create a product like that.

You have my due respect.

Now, a question. How much drag force will be applied in your guess? Will the wing want to roll back putting a lot of stress on the deck? Seems like the forward mounts might not be pushing down all that much. I'm a real novice at this kind of aero engineering.

Edit: I was tpoing this as the above posts were posted. I see the question has already been addressed.

[911pcars]

Originally posted by Jack Olsen
Yeah, that's my plan. The downward force actually does what you're describing -- it pushes down onto the chassis underneath the decklid. But the way the wing extends back means that downward pressure on the wing also pulls up on the lid -- imagine pulling straight back on the wing it wants to lift the forward edge of the decklid. It's like a lever effect.

I'm going to keep the existing mounts, but also integrate in a triangle-shape on the side that folds down below. It will run up a little higher on the rain gutter part of the chassis, but it will also distribute the force more evenly to the decklid itself.

Jack,
The really rigid structure is but 8" below where you are now - the fore-aft longitudinal rails below the rain gutters. The rear engine mounts are close by as well. However, it's a convoluted path from here to there so maybe save this in case you need a plan C.

Sherwood

[Jack Olsen]

Thanks, Milt. Coming from you, that means a lot. But I have to say, when you see it up close, you'll smile, pat me on the back, and be glad that I'm keeping my day job. I love doing this stuff, but I'm not especially skilled at it, and I don't have the patience or the persistence that a real craftsman has.

Then again, maybe my lack of engineering and fabrication experience is why I'm willing to jump into projects like this with both feet.

Or maybe I just have too much free time.

As I understand it, though, the drag issue is minor (when it comes to my decklid and mounts) compared to the way the design of the uprights makes downforce mimic a rearward tug (that lever effect). Smdubovsky illustrated it in a different thread:



I've gotten input on this project from a bunch of guys on Pelican, but also from a couple of other guys. One of them does this stuff for a living, but doesn't want me to share data he's produced based on actual simulations he's run. But the other guy (who I met through the S2000 BBS) did simple tests on the airfoil for lift and drag. The drag numbers would be kind of lousy if I was designing an airplane, but aren't going to pose a structural problem for my uprights or mounts. (Peter Bull, the Pelicanite who worked out this airfoil for me, also looked at drag coefficients when we were comparing airfoils. I don't think we talked about actual numbers, though.) The S2000 guy's charts put the maximum drag at about 29 pounds at 120 mph.

[Jack Olsen]

Originally posted by 911pcars
The really rigid structure is but 8" below where you are now - the fore-aft longitudinal rails below the rain gutters. The rear engine mounts are close by as well. However, it's a convoluted path from here to there so maybe save this in case you need a plan C.

Agreed. I think the stress on the fiberglass was coming from the single bolt in the mount of the rear base piece. I didn't anticipate the way the rear mount would twist from pressure from where it connects to the aft side of the upright. Adding a second bolt might be enough to fix it. And simply tying to the two base pieces together would probably eliminate the problem -- but I'm going to take it a step further for my Plan B.

[Tim Hancock]

Jack, nice job. FWIW, epoxy will bond aluminum when prepped properly (Grumman Yankees, Tigers and Cheetahs are all simply glued together with epoxy as opposed to more traditional rivets).

I see you installed a diagonal round tube to keep the wing from flexing side to side. Round tubes produce a lot of drag and your single small tube will only be effective in preventing flex when in tension. When in compression, it will tend to buckle. This may not be an issue, however I am a bit concerned that you may get a flutter at high speed without a little better bracing to keep the thin side plates from going into a destructive flutter mode.

The drag from your "round" diagonal tube is probably not worth worrying about, but streamlined tubing would be better. If it were me, I might think about adding an additional "jury" strut to give this lonely tube some help in the buckling dept or make your side plates shorter or make them from 4130 streamlined tubing which would make them stiffer yet still relatively drag free. I realize that theoretically you are not introducing side loads, but flutter once initiated is wicked.

These are just friendly observations, I think you did a great job.


OOPS EDIT: After re-reading, It looks like I missed the part where you stated the diagonal had a streamlined cover.

[JohnJL]

Wow, no technical comments to add, but I am seriously impressed by your inginuity and workmanship. Congrats, I'm subscribed!

[jeffgrant]

Wow... very impressive!

I swear, you're making it harder and harder for me to tell you and Jack Ridley apart!

(FYI, Jack Ridley was, among other things, the Project Engineer on the X-1 tests with Chuck Yeager, as well as one of my all-time heroes)

[randywebb]

this will make a nice new product for Olsen Enterprises, Inc.


- eventually you may want to move the ducktail to a different location