9 Lives Racing

Occam’s Racer Endplate Test

In this article I take a deep dive into endplates: why we use them, various design considerations, and then testing different endplates in a wind tunnel. I asked several endplate manufacturers to play ball. Some declined, some ignored, but I was able to secure a good batch for testing, and then borrowed or made what I couldn’t acquire. Across my various trips to the wind tunnel, I’ve now tested over 20 endplates, and this article is the first time I’ve put all that data into one place.

Because these tests cover a few trips to the wind tunnel, I’ll list details about the car and wing each endplate was used on. I always use a standard 12×9 endplate as the basis for the comparisons, and I try to be as scientific as possible.

There’s a lot to cover, let’s get started.

Wings need endplates
You won’t read this in a magazine
Endplates by the inch
144 square-inch shapes
Aviation-inspired experiments
CFD-designed endplates
9 Lives Racing endplates
Yeti Racecraft endplates
Miscellaneous endplates
Conclusions
Side force and center of pressure

Wings need endplates

Endplates improve a wing’s performance, but there are other reasons we use them.

Wing performance – According to Simon McBeath (Competition Car Downforce), wings without endplates lose up to 30% of their performance.
Stability – Moving the center of pressure rearwards makes the car more stable.
Structure – Endplates can be incorporated into the structure of the wing itself, and in a dual-element wing, provide the mounting points and adjustments for the upper wing.
Gamesmanship – Endplates hide the airfoil shape; it’s always good to keep your competitors guessing. Fancy endplates also distract competitors from more important things on your car.
Personalization – Endplates look cool, which is a minor point for some, but the entire point of aero, for others.

In this article I’m mostly concerned with point 1, and to a lesser extent point 2. So let’s jump directly into how endplates affect car performance.

Wings create downforce via the pressure differential between the upper and lower surfaces. At the wing tips, that pressure differential creates a vortex as the car moves forward, because the high-pressure top side wants to move around to the low-pressure bottom side.

Endplates provide a physical barrier, holding the air at the ends of the wing, rather than letting it spill over the sides. This fools the air into thinking the wing is longer (a higher aspect ratio). Larger endplates are better at blocking the air, and theoretically should provide better performance.

You can calculate the benefit of a simple rectangular endplate on a 2D wing using the formula AR = AR_actual (1 + (1.9 * h/b)). Mathematically, this means that aspect ratio (AR) is proportionate to the size of the endplate, or in plainer terms, the larger the endplate, the more it reduces drag and increases downforce.

In reality, I haven’t found that the math works out. In the wind tunnel, changing the size of a rectangular endplate doesn’t get me both drag reduction and more downforce, I get more of both or less of both.

That math formula is for rectangles, and rectangles are boring. Endplates can, and should, be sized and shaped to a specific airfoil. Every airfoil has a different shape and area of its high- and low-pressure regions, and most motorsports wings have a low pressure zone that extends very low and in front of the wing.

The low-pressure region extends in front of and below the wing. S1223 shown.

As you probably know, the low pressure side is much more important than the high pressure side. Wings make about 70% of their downforce from suction, and so it’s the underside of a car wing that matters the most. This is why endplates extend further below the wing than they do on top, and it follows that deeper endplates should make more downforce.

To gain even more downforce, you can add a small Gurney flap to the trailing edge of the endplate, to kick air sideways. This makes the wing seem like it has more span. You can also bend the endplate such that some air is deflected upwards, which creates downforce.

In addition to the size, shape, and bends on the endplate, you can add vents. Vents are usually in the form of slits or notches cut in the top of the endplate, or at the trailing edge. Although sometimes you see vents on the bottom of the endplate, mixing outside air into the wake. Regardless of where they are, the point of venting is to bleed some of the high pressure air to the low pressure side. This may seem counterintuitive, because lowering the pressure differential reduces downforce. But, if done cleverly, venting can reduce the wingtip vortex, and thereby reduce drag, while not losing too much downforce in the process.

In a nutshell, or perhaps a better idiom is, in this shell game, that’s what endplate design is: trying to get something for nothing. Either you’re trying to get drag reduction without losing downforce, or trying to get downforce without increasing drag.

So how do these different shapes and sizes combine with cuts, vents, Gurney flaps, and bends? Well, that’s what we’re here to find out. Before we dive into it, here are a few things to keep in mind:

All values are listed at 100 mph. The A2 wind tunnel operates at 85 mph, but I use 100 mph as a standard convention (as does Simon McBeath, AJ Hartman, and other people smarter than me).
A gallon of gas is about 6.2 lbs, and so if you see an endplate that makes that much additional downforce, an extra gallon of gas in the tank is about the same amount of additional grip.
One horsepower is about 7 lbs of drag (on the cars I measured here), and so an endplate that reduces drag by say 3.6 lbs gained about .5 hp at 100 mph, which is negligible.

You won’t read this in a magazine

I think I’m the first person to publish an exhaustive endplate test with wind tunnel data. It begs the question, why me? Or rather, why haven’t I read about this already in a magazine? Car magazines will dyno test different air filters, intakes, exhausts, motor oils, etc. They’ll rent track time to test a new 200TW tire as soon as it comes out. But aero testing? Notsomuch.

But people also want to know about aero, and one of the most intriguing items has got to be endplates! They are cheap, easy to install, and come in a variety of shapes and sizes. Most endplates are “CFD designed” and if we believe the wild (and unsubstantiated) claims of increased downforce and/or drag reduction, they are a great bang for the buck. Which is best?

I can’t say for sure why car magazines won’t test things like this, but it seems like a combination of the following:

Lack of interest – If you search the archives in GRM, Road & Track, or any other rag outside of Racecar Engineering (which stands head and shoulders above everything else in this regard), you’ll see precious few articles on aerodynamics. I can count them on one hand.
Conflict of interest – Manufacturers place ads in magazines, which puts food on the table. Comparative test only works out for the winner. So if Brand X is a regular advertiser and comes in last place, that’s difficult for the magazine. In the end, it’s probably better to write articles on vinyl wrapping, or HID headlights, or whatever else passes for motorsports these days.
Lack of knowledge – Most car magazines buy parts off the shelf, investigate nothing, and blindly trust manufacturers claims. On the Dunning-Kruger of Aerodynamics, car magazines are firmly in the FanBoi stage: high confidence, low knowledge.

So that’s why you’re not reading about endplate testing in a magazine. But at the price the aftermarket charges for a piece of aluminum, consumers deserve to know what different endplates do for drag and downforce, and ultimately, if they are worth it.

I guess the next question is, if not them, then why me? Why have I taken it upon myself to be the Consumer Reports of aerodynamics? Blame BBTI.

This would be a great time to hit the Buy Me a Coffee link, which supports true investigative journalism, and actual wind tunnel data. The money all goes back into more wind tunnel testing, which results in more articles on this site. Thanks for the support, let’s keep this going.

Endplates by the inch

One of the websites that has inspired my aero research, and this endplate test in particular, is Ballistics by the Inch. BBTI tests the velocity of different pistol and rifle bullets from different length gun barrels. They start with a long barrel, and then progressively cut 1” off it, fire a bunch of rounds to measure and average velocity, and repeat. When the barrel is reduced to a 2″ stub, they move onto another caliber and start over with a full length barrel.

The results show that some calibers, like .223 and 357 Magnum, get much better with a longer barrel, while some, like 38 Special and 45ACP, don’t. This is all down to the propellant (gunpowder) ammo factories use, so you could handload for different results, but it’s still great info for anyone buying off the shelf ammo.

Key takeaways are that short-barrel firearms using slower burning powders produce more flash and noise, but not a lot more velocity. If you have a snub-nosed 357 Magnum, you’re better off feeding it 38 Specials. Conversely, slower burning powders are better in longer barrels. So if you have a 6” 357 Magnum revolver, it has twice the muzzle energy of a 3” barrel of the same.

Good stuff. Useful knowledge. They’ve tested 25 calibers, real firearms vs bench barrels, cylinder gaps, and rifling. The ammo cost alone must have been staggering. Who are the people behind these tests? Are they ammunition manufacturers? Guns and Ammo journalist? Nope. Just a couple of guys who did it for science. They were willing to pay for crates of ammo, destroy a lot of barrels, and spend hours on laborious experiments.

I was doubly inspired by BBTI, not only for doing the Lord’s work on their dime, but also by the whole “let’s keep cutting it shorter and see what happens” methodology.

I wanted to see what would happen if I did that with endplates, but in reverse. So that’s what these first endplate tests are: start with a small rectangular endplate and get progressively larger. I don’t have the time (money) to dick around with 1″ increments like BBTI did, so I decided to make end plates at 3″ intervals: 6″, 9″, 12″, 15”, and 18”.

I didn’t start with a 3″ endplate, as that would have been only half an inch depth under the wing, and that seemed silly based on the size of the low-pressure region. And I didn’t go larger than 18″, for a couple reasons.

I use recycled street signs, and anything over 18″ long means cutting into a 24″x24″ or larger sign, which is wasteful.
I felt that if I extended the endplate over 18″, they might bend inward at the bottom. The aluminum signs I use are quite sturdy, but I’ve seen this bending on my buddy Bronson’s car, and it looks like anti-performance.
Very large endplates are going to have considerably more side force. Meaning that they would move the center of pressure rearward, and that might have a greater effect on handling than whatever the endplate is doing to the wing’s performance. That’s a separate topic, and difficult to discuss without knowing the weight balance and silhouette of the car in question.

Simon McBeath published a similar test in Racecar Engineering, using CFD. He started with a small rectangular endplate and then increased the depth of the endplate by 50mm each time. He stumbled upon magical in-between size (125mm IIRC), which outperformed all the others. Keep this in mind, because it kinda happens later in the test, as well.

I won’t republish Simon’s results here, subscribe to Racecar Engineering and you can find that article in their archives. If you want to cross-list my data with Simon’s, note that Simon measured the depth below the wing, and I use the total endplate height. My endplate depth is about 2.5″ less than the total height, so a 6″ end plate has about 3.5″ of depth.

Note that I did the rectangular endplate test on my Veloster N with a 71″ Wing-Logic MSHD. I used reverse swan neck wing mounts mounted at roof height, with the usual quarter-chord overlap over the rear hatch. I feel the wing was high and wide enough that the body shape didn’t make much difference on the endplate performance, but it’s always worth noting that the car influences the wing and endplate.

The following table is listed in pounds (I’m embarrassed that I can’t think in metric; I’m too old to make the switch). Positive values are bad, negative values are good: less drag or less lift (= downforce). I list the total Lift, which is the lift at the front (Lift F) and rear (Lift R) added together. Rear downforce and drag can lift the front of the car, and it’s interesting to see that relationship here. I’ve also included a column for the L/D ratio the car, and how that changed with each endplate.

Endplate	Drag	Lift	Lift F	Lift R	Car L/D
12×6	-0.5	1.5	0	1.5	0.88
12×9	0	0	0	0	0.88
12×10.5*	1.1	-0.8	0.35	-1.15	0.88
12×12	2.2	-1.6	0.7	-2.3	0.88
12×15*	4.7	-2.55	1.15	-3.7	0.88
12×18	7.2	-3.5	1.6	-5.1	0.87

Rectangular endplates.

Let’s take a look at that results in detail:

12×6 – The smallest rectangular endplate reduced drag by .5 lbs and lost 1.5 lbs of downforce. Not a great tradeoff, but also these are tiny numbers, and if you have 6″ endplates now, there’s no reason to “upgrade” them.
12×9 – This is a street sign cut in half, and is the endplate I put on every wing, to baseline the testing of other endplates. Its values are all zero, because everything else is being compared to this one.
12×10.5* – I didn’t test this size, the values in the table are an average between 9″ and 12″, which is a fair assumption given the data. I included this in-between size because it’s the same dimensions as the 9 Lives Racing Megga endplate, which, if we believe the advertisement, is supposed to increase downforce by 10%. I’ll be kind to 9 Lives Racing and compare the Megga-sized endplate to the 6″ rectangle, and you can see the larger endplate increased downforce by 2.3 lbs. Given that the wing is making about 260 lbs of downforce, this is less than 1% more downforce than a 6″ endplate. Claiming 10% more downforce is 1000% wrong.
12×12 -This is the maximum area allowed in NASA Super Touring, and I’ll go into more detail about that in the following section. It makes 1.6 lbs more downforce than the 9″, and trades that for 2.2 lbs more drag. These are nothing numbers, which is why the L/D ratio of the car didn’t change.
12×15* – I didn’t test this size for financial reasons, the values in the table are an average between 12″ and 18″. This seems a fair assumption based on how little things change, and there’s no reason to use this size IRL.
12×18″ – This is a full-length street sign, and made the most downforce, but also added 7.2 lbs of drag, which is equivalent to losing about 1 hp at 100 mph. The endplates bent inward from the suction below the wing, and this can’t be good, because this is the only endplate that changed the L/D ratio of the car, and it did so in the wrong direction. The A2 wind tunnel operates at 85 mph, and so at a higher speed, you’d see even more bending, and probably less efficiency.

You can see the 18″ endplate bending inward at the bottom. Some of this may be influenced by the location of the wing and body shape. For aerodynamic efficiency, the 18″ endplate was worse than all of the others.

After reviewing all of this information, it appears that every 3″ of depth adds about 1.5 lbs of downforce, but it also adds proportionately more drag. If we remove the 18″ endplate from the test, I can conclude that different sized rectangular endplates result in the same L/D of the vehicle.

In my testing, I found two exceptions to the above statement:

We tested a 300×300 mm endplate (close enough to 12×12) endplate on Raul’s BRZ with a 68″ PCI wing, and the large endplate lost over 20 lbs downforce compared to a standard-sized flat rectangle. For whatever reason, on this wing, on this car, bigger was not better; it was quite a bit worse. I didn’t include this data in the preceding table, because I didn’t test my street sign endplate on his car, but something a little different in size. Anyway, avoid this size on a PCI wing.
In this experiment, I didn’t find a rectangular endplate that was too small, but later in this article, the trapezoidal Wing-Logic endplate also underperformed compared to a 6″ rectangle.

Both of these were outside the Goldilocks zone of “not too big, not too small” and if I were to make a rule of thumb about rectangular endplates, I’d say make them about a chord in height. If you have a 9″ wing, then a 9″ tall endplate is about the right size. Unless you stumble upon some magical size, then going significantly over or under a chord’s height could be a lot worse.

300×300 endplate was too large and oddly lost a lot of downforce compared to a smaller endplate.

144 sq-in shapes

NASA Super Touring rules specify a 144 square inch limit for endplates. Go over that size, and your car is illegal. The rule leads people to believe that bigger is better, but as we just saw on Raul’s PCI wing, bigger is sometimes a lot worse.

It begs the question, why did NASA create this rule? If you look at the rest of the rulebook and the aero values, it’s pretty obvious that guessing at aerodynamic performance, and not measuring it, is their modus operandi.

But let’s put that all aside for a moment and say that you want to maximize your endplate size for the NASA ST rules, the easiest way to do that is a square 12″ x 12″ endplate. But there are other size rectangles that result in the same 144 square inches. Indeed, there are other shapes that result in the same area, but that extend further below the wing. Theoretically, this should produce more downforce. Maybe we can get something for nothing?

I drew up a few endplate designs and sent them over to Tim Miller, and he suggested I try a delta shape with a long axis of 16″. This came out to 142 square inches, but close enough to 144.

I compared C to E: similar area, different shape.

Now you might be wondering which way the end plate faces? It’s a good question, and so I tested it both ways, delta forward (with the taper facing forward) and delta rear (with more surface area forward).

Note that the car, wing, and position of the wing are all the same as in the rectangular endplate tests. But this time I’m using the 12×12 as the basis, because it’s all about the max NASA size.

Endplate	Drag	Lift	Lift F	Lift R	Car L/D
12×12	0	0	0	0	0.88
Delta forward	2.1	-1.6	-0.2	-1.4	0.88
Delta rear	1.9	-2.5	0.4	-2.9	0.88

144 square inch shapes compared

12×12 – This is the basis for the comparison, and so all its values are zero.
Delta forward – The endplate was placed such that the tapered edge faced forward, with more surface area to the rear. This made 1.6 lbs more downforce than the 12×12, but also more drag. End result, no discernable difference.
Delta rearward – The same endplate flipped around so that it had more surface area forward, made more downforce. This makes sense, since much of the low-pressure region is ahead of and below the leading edge of the wing. What’s surprising is this direction also had less drag than the other way around. I’ve been saying this for years, that endplates should be designed with more surface area forward, and I feel vindicated with this evidence. But also, these are such small numbers that it didn’t change the L/D ratio of the car, and whatever point I was trying to prove is pointless.

At this juncture I can conclude that that any rectangular endplate from 6″ to 12″ tall, and probably any 2D shape, will result in the vehicle having the same aerodynamic efficiency. Unless you go too small or too large, or somehow luck into a size that has magical abilities, endplates of different sizes and shapes are just tiny tradeoffs between drag and downforce, and all of them amount to the same aerodynamic efficiency.

So if all 2D shapes and sizes are the same, what about 3D shapes?

Aviation-inspired experiments

I wanted to try a couple different endplate ideas based on things I’ve seen on aviation wings. These tests were on the Veloster with the 71″ MSHD, reverse swan necks, etc.

Airfoil shape – The first experiment was to use an endplate that’s shaped like an airfoil, with a thick leading edge, tapering back to a sharp trailing edge. I made this out of foam and covered it with tape, and then stuck it to a 12×6 endplate. I eyeballed the shape, it was probably close to a Clark-Y. Theoretically, this should work better when the wing is in yaw (going around a corner), as the rounded leading edge should have better flow attachment, and the shape of the airfoil should kick air a little more sideways, making the wing seem like it has a higher aspect ratio.
Fillets – The other thing I wanted to try was to reduce the intersection drag between the endplate and the wing. You see this on airplane wings, at the wing root. That 90-degree corner where they meet is a place where flow separates, and radiusing that area should help flow attachment. I had my friend Alyssa 3D print me some plastic fillets with a radius.

Rounded fillet idea was better on paper than in the wind tunnel.

While both of these ideas sounded great in theory, neither of them worked out.

Endplate	Drag	Lift	Lift F	Lift R	Car L/D
12×6 plain	0	0	0	0	0.88
12×6 airfoil	-1.3	26.8	8.8	18	0.79
12×6 fillet	-4.8	24.7	9.7	15	0.81

12×6 – This is the plain rectangle I used in the endplates by the inch test. All of its values are zero, because the other endplates are based on this.
Airfoil – Making the endplate thicker and shaped like an airfoil reduced drag, but it also lost 26.8 lbs of downforce compared to the unadorned street sign. The aerodynamic efficiency took a dive, down to .79. Ouch.
Fillet – The fillet reduced drag even more, but downforce went down by almost 25 lbs. For a low-drag application like an airplane, where maybe lift is not the primary concern for long distances, perhaps this is a good idea. But for a car wing, not at all.

In retrospect, the airfoil shape was probably too thick, and reduced the effective length of the wing by 3″, which was the thickness of the airfoil, and about 4.5% of the span. Total downforce went down by 11.5%, which means there was still an unfavorable tradeoff between drag reduction and downforce. The airfoil shape would be most effective when placed in yaw, but I didn’t test that. There may be some benefit to thickening and rounding the leading edge, but not at the expense of reducing the area under the wing!

The fillet under the wing also shortened the effective wing length, and that’s probably why it lost downforce. It’s a similar problem to the airfoil shape. If I redesigned this, I’d radius only the trailing surfaces after the thickest part of the wing.

But neither one of these is a game changer, and in fact both were utter crap. The theme of “you get drag reduction or downforce, but not both” remains true.

CFD-designed end plates

My guesses at improving endplate design were interesting but not fruitful. Taking my ideas and shaping them into reality, and then testing them in a wind tunnel, is a slow and expensive process. People smarter than me do their development in CFD. This allows them to try many different iterations on the computer, and arrive at an acceptable tradeoff between drag and downforce.

But there are three common problems I see with the CFD approach.

Testing – Nobody seems to test their CFD-designed endplates in a wind tunnel. They think (or rather, want you to believe) the CFD is valid, when in fact it’s never been calibrated by putting it into a feedback loop. “CFD-proof” isn’t proof; if you see those words you are being scammed.
Free stream – Most of the CFD I’ve seen is done in free stream. The fact is, endplates should be designed not only for a particular wing, but for a specific car, with an exact wing size and wing location. If you test wings in free stream, meaning that the wing is suspended in air, but not attached to the car, you can absolutely design the most efficient endplate for any given wing. But when you put the wing on the car, the game changes entirely. Developing endplates (or any aero device) in free stream CFD and applying that to all cars is utter bullshit.
Wing efficiency vs vehicle efficiency – Cars are large and draggy, and have shitty aerodynamics. In order to improve the L/D ratio, you need to make huge changes to downforce, not small changes to drag. Adding more wing angle makes the wing less efficient, but it makes the vehicle more efficient.

But if CFD isn’t real, the wind tunnel isn’t real either. In the real world we have winds from every direction, cars in front of us, and in the situations where when we want the most downforce, in a corner, the car is in yaw.

I didn’t test those things, and they are certainly important. So as much as I might malign CFD for giving us false hopes, I’m also acknowledging that wind tunnel testing isn’t 100% accurate, either. We are all doing the best we can, with the information we have available.

I tested CFD-designed end plates from Yeti Racecraft and 9 Lives Racing, which I cover below. At the wind tunnel we also tested some CFD-designed top secret endplates I can’t show any pictures of. The results were exactly in line with all of the other CFD-designed endplates, with their fancy vents, cuts, and shapes, and so I’ll wrap up that in the final conclusion.

9 Lives Racing sells four endplates, three rectangles of different sizes and a CFD-designed shape that has a couple cuts and a bend in it.

The basic endplate measures 10” chord by 8” depth. The wing is positioned at the rear of the endplate, so that the trailing edge of the wing is at the rear edge of the endplate. This puts most of the endplate’s surface area in front of and below the wing, which is proper.
The mini endplate looks to be the same as the basic, but with less depth. I didn’t test this one, but I’d guess it performs similarly to the 6” endplate I tested.
The MEGGA!! endplate is larger than the basic in both directions, measuring 12” chord and 10.5” depth. I already covered this size in the Endplates by the Inch section.
The CFD V2 measures 10.5″ chord by 12” depth with a semi-circular arc cut into the leading edge and a vent cut into the top side. It measures about 115 square-inches, well under the silly 144 square inch NASA Super Touring rule.
Years ago there was a CFD V1 endplate, which had the same shape, but without the bend on the trailing edge. Based on the shape of the low pressure region below the wing, this semi-circular cut seems like it would reduce downforce, just the way that the Delta-shaped endplate I tested showed that more area forward was better than more area reaward. Anyway, I tested CFD V1 on a previous trip to the wind tunnel, and I ran it in both directions to test my hypothesis. I’ll include that data in the summary table.

Note that the these endplates were tested on two different cars, my Veloster N with a 55″ 9LR original (Express) wing, and Pete Mink’s Porshe 944 with a 68″ Evo wing. I didn’t include a vehicle L/D ratio column in the table, because I can’t relate it to both cars. In any case, you can probably guess that the vehicle L/D didn’t change much.

Endplate	Drag	Lift	Lift F	Lift R
12×9 rectangle	0	0	0	0
9LR Basic	0.8	-7.2	1.2	-8.3
CFD V2	1.1	-3.7	-1	-2.7
CFD	4.1	-1.8	-0.8	-1
CFD reversed	3.5	-1.5	-0.4	-1.1

12×9 – Street sign cut in half, everything here is compared to this.
9LR Basic – What’s super interesting about this endplate is how well it performed compared to the 12×9. It has 2″ less chord and 1″ less depth, and yet it made 7.2 lbs more downforce. Drag went up, which is what happens when you add downforce, but by less than 1 lb. Nevertheless, this is better performance, and if I had to guess, I think it has to do with a) stumbling upon a “magic” size, and b) positioning of the wing on the endplate. On the street sign endplates, I centered the wing fore and aft, but on this endplate, the wing is as far rearward as you can put it. This endplate shows that a proper rectangle is as good, or better, than most anything.
CFD V2 – Compared to the Basic endplate, the CFD V2 makes less downforce and more drag. Which is odd, because their advertising claims the CFD V2 makes 10% more downforce and 7% less drag, and it’s actually worse in both directions. But maybe they were comparing that to no endplate at all? In any case, the Basic is better.
CFD V1- I tested the first version of the CFD shape on my Veloster, and it made more downforce than the 12×9 street sign, but also more drag. When I reversed the endplate on the wing (the holes didn’t even match up, so the endplate was askew), the numbers were slightly different, but basically the same. This is no surprise, since we saw this already with the Delta-shaped endplates, putting them forward or backward didn’t move the needle. So it’s expected that the same result happened here.

Two things of interest came out of this test, one is that there is sometimes a sprinkle of magic in rectangular endplates, and one size may work slightly better than others. Simon found this in his CFD experiment, and I’ve found the same in this test, both in sizes that work better than they should, and sizes that are quite a bit worse. Now we are only talking about a difference of 7 lbs of downforce at 100 mph, and that wouldn’t change the performance of the car in a way that you’d notice. Still, it’s neat to see this.

The other interesting tidbit is the difference between the CFD original and CFD V2, with the rear bend on it. The rear bend added 2 lbs of downforce and reduced drag by about 3 lbs, and is an obvious improvement. While these are tiny numbers, they are at least going in the right direction. But the Basic rectangular endplate still outperformed both V1 and V2.

Yeti Racecraft

Yeti Racecraft is an aftermarket manufacturer making CFD-designed endplates and other aero parts. Yeti’s Brian Kemerley designs each one specific to an aftermarket wing. Meaning that Brian’s endplate for a PCI wing is a different size and shape as the one he makes for 9 Lives, which is different from the one he designs for Wing-Logic. This makes sense because the pressure distribution is going to be a different size and shape on different wings.

But the endplate shapes aren’t hugely different, they all have the same family resemblance, with more surface area forward than aft. A rectangular endplate forms a vortex on the corners, but Brian’s “swoosh” shape merges two vortices into one, which results in less drag. There are vents cut into the top, to bleed some high pressure air to the low pressure side, and a long tail with a notch in it, presumably to do the same.

I tested this model, both with and without the rear flick at the trailing edge.

Some of the Yeti endplates are flat, and some have a small Gurney flap on the trailing edge. Unlike the 9LR CFD V2 endplate, the Yeti bend is only on about half the trailing edge. In the data below you’ll see that this worked, although I wonder how it would have performed if the lower half had the flick on it, rather than the upper half.

These endplates were all tested on a Porsche 944 with a 68″ 9 Lives Racing Evo wing and their standard wing mounts. As a side note, I helped Pete install the wing mounts and I found them to be well designed and constructed.

Endplate	Drag	Lift	Lift F	Lift R	Car L/D
12×9 rectangle	0	0	0	0	0.64
Flat swoosh	-0.9	-0.1	-0.3	0.3	0.64
Flick swoosh	0.2	-4	0.5	-4.6	0.65
Time Attack	5.9	-16.7	5.8	-22.5	0.69

12×9 – Same rectangle, baseline for others.
Flat – The swoosh shape, topside vents, and rear notch combined to do… not much. Compared to the basic 12×9 rectangle, the fancy shape had 1 lb less drag and .1 lbs less downforce. These are barely measurable figures, and no better than a generic rectangle, and worse than a 9LR basic.
Flick – The same shape with a half-length Gurney flap gains 4 lbs of downforce and .2 lbs of drag. That’s so little drag that we can ignore it, and call it a 1.5% gain in downforce for free. Well, not exactly free since the endplates cost $175, but this is an actual performance gain, not fantasyland. But still the 9LR Basic was better.
Time Attack – This endplate has a much larger top vent and a radical double bend at the bottom edge, which scoops air almost like an outboard wing. It made 16.7 lbs of downforce and added 5.9 lbs of drag. That’s a 2.8:1 L/D tradeoff, which is remarkably good. This is slightly better than the L/D you see when adding Gurney flap to a wing.

Yeti Time Attack endplates have a large scoop at the bottom, almost like a mini wing.

I have a knee-jerk reaction against CFD designed endplates, because most of what I’ve seen is an expensive product, supported by misleading claims, and a zero-sum result. But Yeti Racecraft has made at least one endplate that proves Brian can design a CFD endplate that outperforms everything else.

One nit: the description on the Yeti site says this endplate also reduces drag, and while that may be true in the CFD, it’s not true in the wind tunnel. No endplate adds downforce and reduces drag. But what is true is that this is an excellent endplate, that results in better vehicle efficiency than any other endplate I tested. Bravo.

Yeti’s CFD shows how effective that bottom bend is, effectively making the wing seem like it has more span. That it does this with such a favorable L/D ratio is astounding.

Yeti also makes a hybrid shape that combines their basic swoosh shape with a less aggressive bottom bend than the Time Attack. I didn’t test this one, but it probably sits somewhere between their Time Attack and swoosh-flick.

I didn’t test this one, which combines the Yeti swoosh shape with a less aggressive Time Attack bend.

At the end of this article I give each endplate a grade, and the only thing keeping the Yeti Time Attack from getting an A+ is a couple adjustment holes. With an endplate like this, when you adjust the angle of attack on the wing, it also adjusts the angle of the “winglet” on the endplate. So I’d like to see adjustment holes here, so that the endplate can be pivoted a few degrees fore and aft.

Miscellaneous endplates

I’m going to add three more endplates to this discussion, the first is the endplates that come in the box when you buy a Wing-Logic MSHD, the next is the basic 12×9 street sign with a bottom vent, the last is an endplate I made back in 2019.

The Wing-Logic endplate is a 10×5.5 parallelogram. This is the smallest endplate in the test. It doesn’t come pre-drilled, and so I placed the endplate about where I placed the rectangular endplates, with most of the area below the wing.

The next one is the standard 12×9 street sign, with a single slash cut into the bottom of the endplate, and then bent such that it directs air from the outside of the endplate into the wake behind the car. The purpose of this is to (theoretically) reduce the trailing vortex and reduce drag. There’s no picture of this one, not because it’s a secret, but because you don’t want to do it.

Finally my Occam’s Racer endplate was one I made back in 2019. I shaped it so that there’s more area forward and less at the trailing edge. I put a notch in the top side trailing edge, because I’ve seen that on other endplates, and I bent a small 90-degree Gurney flap into the trailing edge.

The endplates were tested on different cars and with different wings, but I did baseline the 12×9 endplate on every comparative test, and so we can have a valid discussion about the results.

Endplate	Drag	Lift	Lift F	Lift R
12×9 rectangle	0	0	0	0
12×9 vented	2.8	-2.4	.8	-3.2
Wing-Logic	4.2	11.3	7	4.3
Occam’s Racer	6.9	-8.9	0.8	-9.6

12×9 rectangle – Baseline for all runs.
12×9 bottom vent – I thought the bottom vent would reduce the wingtip vortex, thereby reducing drag, but it actually did the opposite. By adding outside air into the wake downstream, it energized the flow and downforce went up by 2.4 lbs, which surprised me. However, drag went up by 2.8 which is the usual tradeoff, and not a very good one, at that.
Wing-Logic – This is a really weird and useful result. The total area is 55 square inches, which is not that much smaller than a a 12×6 rectangle at 72 inches, which you may recall performed about the same as a 12×9. However, the Wing-Logic endplate lost 11.3 lbs of downforce and gained 4.2 lbs drag. Ergo, this endplate is undersized, and doesn’t block enough of the high- and low-pressure regions, resulting in losses both ways.
Occam’s Racer – We’ve already seen that shapes are basically meaningless, and I doubt that top notch does anything, so the 9 lbs of downforce is probably down to the Gurney flap alone. This comes with a drag penalty, and the resulting 1.29 L/D ratio would only be beneficial on a car with lousy aerodynamics or at low speed.

The key takeaways from these tests are that there is an endplate size that is too small, and we found it. We also found out that bottom vents are a zero sum modification, just like most everything else. While the Gurney flap worked to increase downforce, I think it was too aggressive and too vertical. Angling the flick upwards, like Yeti does, makes more sense.

Conclusions

An endplate can be designed to reduce the strength of the wingtip vortex, and thus reduce wing drag. This can be accomplished by cutting vents in the top of the endplate, or using a 3D airfoil shape, or radiusing the intersection between wing and endplate, or shaping the endplate to combine smaller vortices into one. From my testing, I can say that anything that reduces drag also reduces downforce.

Conversely, an endplate can be designed to increase downforce. This can be simply a larger endplate, or sending more air sideways using a Gurney flap on the trailing edge, or mixing outside air into the wake, or by deflecting air upwards. From my testing, I can say that anything that increases downforce, also increases drag.

Combining the drag reduction tricks with the downforce producing tricks is a zero sum game. Of the endplates I’ve tested, there is no endplate that both reduces drag and also increases downforce. None. Most modifications to the size, shape, vents, etc., all result in a vehicle with the same L/D ratio.

However, that doesn’t mean the performance would be exactly the same. You might want to choose the endplate that has the least drag for a high-speed track, and the most downforce for a low-speed track. It’s a very minor consideration, but it’s worth noting that overall vehicle efficiency doesn’t mean it’s the same performance. Although what you’re really after is better vehicle efficiency.

Because cars are large complex objects with terrible aerodynamics, endplates that increase downforce usually results in the vehicle having a better L/D ratio, even though the downforce-producing endplate adds drag. For example, if your car has a L/D of 1:1, then any aero modification that is better than a 1:1 L/D ratio is going to improve the aerodynamic performance of your car.

Of the endplates I tested, only the Yeti Racecraft Time Attack endplates improved downforce and vehicle efficiency by a meaningful amount. As such, the Yeti Time Attack endplate is the clear winner of this test. Everything else is meh, or worse.

And here’s where a real opportunity lies. If you replace the Wing-Logic endplates with the Yeti Time Attack, you’d get something like 28 pounds more downforce for not even 2 lbs of drag. This combination should be offered out the door, because that’s about 10% more downforce for almost no penalty to drag. Wow. Loot. Wing-Logic and Yeti Racecraft need to get together and offer this peanut butter and jelly sandwich.

Other than combining this worst-of case with a best-of case, there isn’t a lot to recommend one way or the other. These are tiny numbers, people. For most of us, a plain rectangle one chord in height is going to perform as well as anything else. Any amount of time and money spent developing something better than a medium-sized rectangular end plate is better spent doing anything else, anywhere else on the car. Spend this time, money, and energy on your spouse.

At the highest level of motorsports, I’m sure it’s possible to design an endplate that works with a with a particular wing, on a specific car, at an exact angle, height, and setback distance. With those variables locked down, maybe there’s a minuscule gain in performance? But on all the cars, wings, and endplate combinations I’ve tested (and some data I can’t share), the different shapes, sizes, cuts, vents, etc., do nothing for the performance of my car.

There are a few things I didn’t test, or rather would re-test, given unlimited funds. As it is, I don’t plan to test endplates ever again (NEVER AGAIN – I’m shouting). But if I did…

Position – Placing the wing further rearward on the endplate, such that the trailing edge of the wing and trailing edge of the endplate aligned, seemed to be better. Whether this was because there’s more surface area in front of the wing, or whether it’s because there’s less surface area after the wing, I don’t know. Flying in the face of that generalization are the Yeti endplates, which all extend rearward a good amount, and they were all pretty decent. I should have cut one of the Yeti endplates right at the trailing edge to see what happens. But if there are gains to be had, they are likely to be pretty small.
Yaw – Race cars go around corners, and in doing so, typically have around 5 degrees of yaw. As a result, thin and flat endplates will experience some degree of flow separation at the leading edge. 3D endplates, with thick radiused leading edges would seem to be better in this regard. I tested thicker endplates, which were not good, but perhaps in yaw, they would have been? Maybe with a different airfoil profile? That mystery remains.

In the end, endplate performance doesn’t matter much. So perhaps the least important performance aspect is the most important purchasing decision: how the owner feels about it. If you like the look of a certain endplate, buy that endplate. If you’re a fanboi of a particular company, buy their endplates. And if junkyard-chic appeals to you, go to a scrapyard with $2 and cut a street sign in half.

Side force and center of pressure

Before I wrap up this article I want to discuss one more datapoint that I included in the final table of results: side force. This value ranged from -1.7 lbs to +5.7 lbs, with larger endplates having a greater side force, as you might expect.

I can’t tell you how much side force is going to affect the handing of the car; all cars are different, and I haven’t experimented with it myself. But any surface area at the rear of the car moves moves the center of pressure rearward. Station wagons and hatchbacks, with their large cargo areas, and LeMans prototypes, with their shark-fin bodywork, all move the center of pressure rearward.

Moving the center of pressure rearward changes how the car handles. The distance between the center of gravity and the center of pressure is called the static margin. From a driver’s perspective, a low static margin means the car gets looser at high speed. Conversely, a higher static margin means more high speed stability, and fewer corrections when the car goes over the limit. But the tradeoff is a car that is harder to steer, or may in fact understeer at high speed. Since wings are at the very rear of the car, the size of the endplates can have a significant effect on the static margin.

On a mid- or rear-engine car, the center of gravity is rear biased, and the static margin is low. Using larger endplates is a way to increase the static margin, and gain some high speed stability. If I had a 911, I’d use big-ass endplates.

Conversely, on a FWD hatchback, there’s a lot of weight up front, and a lot of surface area on the rear of the car, which results in a very large (too much) static margin. In this case, I’d use smaller endplates, because I don’t want to make a bad situation worse.

So when you look at the summary table, you might use the side force value and the weight balance of your car as another variable that factors into your choice of endplates. We already know that most endplates aren’t doing much to affect the car’s performance, and so side force might prove to be more important than a couple pounds of drag or downforce, this way or that way.

I’ll leave you with this summary table and a subjective performance grade. The price of the endplates isn’t factored into the grade, or you’d see a bunch of street sign trash at the top of the list.

Endplate	Grade	Drag	Lift	Side	Cost
Yeti Time Attack	A	5.9	-16.7	-0.6	$200
9LR basic	B+	0.8	-7.2	-0.5	$72
Yeti flick	B	0.2	-4	-0.7	$175
Yeti swoosh	B-	-0.9	-0.1	0.3	$150
Occam’s Racer	B-	6.9	-8.9	0.2	$10
9LR CFD v2	B-	1.1	-3.7	0.9	$240
12×6	C	-0.5	1.5	-0.7	$2
12×9 basic	C	0	0	0	$2
12×9 bottom vent	C	2.8	-2.4	1.7	$2
12×16 Delta R	C	4.1	-4.1	2.3	$4
12×16 Delta F	C	4.3	-3.2	3.5	$4
Megga (12×10.5*)	C	1.1	-0.8	1.35	$198
12×12	C	2.2	-1.6	2.7	$4
12×15*	C	4.7	-2.55	4.2	$4
12×18	C-	7.2	-3.5	5.7	$4
9LR CFD v1	C-	4.1	-1.8	2.2	n/a
9LR CFD v1 rev	C-	3.5	-1.5	1.2	n/a
12×9 fillet	D	0.3	4	1.8	$10
Wing-Logic	D	4.2	11.3	-0.1	$0
12×6, airfoil	F	-1.3	26.8	-1.7	$2
12×6 fillet	F	-4.8	24.7	0.6	$10

If you found this information useful, please Buy Me a Coffee. I’ve spent over $2000 testing endplates in a wind tunnel. The time to build, organize, and write that all up is not included. You may have noticed there are no ads or popups or other annoying shite on my website, because you probably hate that as much as I do. Please support science, open data, and further testing by buying me a cuppa. Thanks!

Back to the A2 Wind Tunnel

My annual trip to the wind tunnel is coming up soon. I had originally planned to do just a single day of testing, but two cars became three, which became four. Now I’m at five. And that’s too many to test in one day.

So I’ll be headed to A2 on 5/18 for hatchback day, then to VIR to instruct for a couple days, and then back to A2 on 5/21 for fastback day. There’s room for a third car on the first day, contact me if you want to get in on this.

Hatchbacks 5/18

The first day will be my Veloster N and Andrew Rivers’ Honda CRZ. We are both testing canards and wings, and some other odds and ends.

I’ve tested canards on my Veloster twice now. The first time I was mostly concerned about the location of the canard. If you bought Wind Tunnel Report 01, then you know that location played an enormous role in downforce. The same exact canard produced anywhere from 11 lbs to 80 lbs of downforce depending on how high I mounted it.

The second time I tested canards I wanted to find out if I was getting downforce from the interaction between the canard and splitter, or was it just the canard itself? I was also able to test a CFD-designed canard from Verus, which didn’t produce a lot of downforce, but also had zero drag.

This time I’m testing canard edge treatment and size. The DIY canards I’ve tested on my Veloster have had a sharp outer edge, but I often see canards with a vertical outer edge, kind of a fence or Gurney flap. So I’m testing two different heights versus a plain edge. Then I’ll also increase the size of the canard to see if more area produces more downforce.

Finally, I want to see if I can get a mid-height canard to interact with the lower canard. On other cars I’ve tested, the lower canard does most of the work, while the mid-height canard seems to reduce the drag of the entire system. I want to see if I can make that happen on my car.

Andrew is doing some different canard experiments, seeing if there’s interaction between various front end components. I don’t want to say too much on this, as those are his tests, but they should be super cool.

Next we are both testing wings. I’m finally getting around to testing a MSHD wing, both as a single and dual element. I’ll bring along a 9 Lives with me as well, since that’s a popular benchmark, and easy to test as a single or dual element.

Because hatchbacks are weird, I also plan to test wing location. I’m fairly certain that higher is better, but I want to check forward and rearward locations, and how those affect front downforce. I also want to test bottom mounts versus swan necks.

Andrew is testing a Simon MacBeath wing with swan necks, and I’m very excited to see how that performs. He’s doing a sweep of angles and a couple different end plates. I plan to throw a MSHD on his car, just to see how they compare.

Andrew’s car looks great with all the aero, and I’m interested to see which hatchback will have better numbers. But no matter how it turns out, we’ll learn from each other’s experiments, and both of our cars can improve from that.

Andrew’s CRZ has very nice swan necks. The tube is a stiffener inside his DIY S1223.

Fastbacks 5/21

This is going to be an exciting day, because we can test wings on proper rooflines, not crappy hatchbacks. Pete is bringing his endurance racing 944, Raul has a BRZ, and AJ is bringing a C7.

As part of the wing tests, we will throw a mess of different end plates on the Porsche. I made an Open Call for testing end plates, but so far not that many manufacturers have joined the game.

Just the same, there are some really cool endplate designs we’ll be testing, with vents, 3D shapes, and other trickery. Even if endplates don’t make a huge difference, they are easy to change out and are visually interesting.

We may try the most radical end plates on two or three different cars, to see if there’s any difference. It’s worth checking out, but I kinda doubt it.

Raul’s BRZ will be the test bed for a few different wings including a PCI, MSHD, APR GTC200, and 9 Lives. I’m also trying to get my hands on a Verus high efficiency wing, and there’s a new nylon wing from Baero that might sneak into the tests if it can get through customs quickly. It has a familiar look, but with a lot more chord, and I have high expectations.

We have to fit many of those wings on the same trunk mounts, which is proving to be a bit of work making brackets. We also want to test wing location, and if setting the wing back a little further is better.

Raul doing the obligatory “is is strong enough?” test.

Raul’s car also has a splitter and so we plan to test splitter tunnels and some other front end stuff.

I’ll write up a report this summer, and probably jump on a podcast to talk about some of the results. Hopefully Kaan from the Blind Apex will come down to see some of testing firsthand, and we can recap that on his show with some special guests.

As usual I’m planning to rope AJ Hartman into participating, but also inviting Ido Waksman and Tim Miller, if their busy schedules permit. Michael Jui from Wing-Logic is also coming for the first day, and so it will be cool to finally meet him in person.

All said, it looks to be quite the party, and we are set to learn a whole bunch about our expectations, and that aero results are often not what you think. Subscribe to the blog if you haven’t already, you don’t want to miss the articles coming out of these tests!

Wing Logic Dual Element

In the previous article, I mounted a Wing Logic wing to Steve Leo’s WRX. He’s been pretty happy with it, but I’ve been wondering about adding more rear downforce. An excess of rear downforce can make a car boring to drive, but it also improves braking, high-speed stability, and requires fewer corrections when you lose control. So while more rear downforce might ruin the aerodynamic balance, if the driver goes 2 seconds faster, let’s call that a better car.

If you want more downforce from a Wing Logic wing, you can increase the size of the Gurney flap by duct-taping down a piece of angle aluminum butted up against the built-in 1/4” Gurney flap. And with the larger wicker, you can add a little more wing angle. But you’ll soon hit a point of diminishing returns, and shortly after that, the wing will stall out. I haven’t run a full sweep on this wing, but I’ll guesstimate that a Gurney flap 1″ tall (very draggy) and an angle of attack around 11-12 degrees is the limit.

If you still need more downforce, the easiest way to do that is add a second element above the main wing, set somewhere between 25-35 degrees angle of attack (in relation to the main wing). Unlike the single wing, the double wing won’t stall because of the slot between the wings; Air shoots through the gap at great speed and this keeps air attached to the underside of the upper wing. Thus you can run more angle on the upper wing, which effectively increases both the chord and camber of the entire wing, without flow separation.

If you have a 9 Lives Racing wing, they can sell you a dual element wing for around $440 (with shipping). The kit includes the upper wing, plus adjustment brackets that go inside the standard end plate, plus little brackets that go in the Gurney flap slot. It’s a clever arrangement that’s easy to install and remove. I tested the double wing in a wind tunnel, and came away really impressed with how well it worked.

If you have a Wing Logic wing and you want a dual element, you’re shit out of luck. There isn’t a similar kit available, and I have yet to see anyone cobble something together. I wonder if the reason for that is because some people believe (incorrectly) that you can’t put a dual-element wing on top of a wing that has a Gurney flap? There was a recent discussion of this on the Professional Awesome Facebook group, and it seemed like most of the people said you should cut off the Gurney flap, or a dual wing won’t work with a Gurney flap on the lower wing, or that Gurney flaps only work on the top wing. That’s horseshit.

You can absolutely put a Gurney flap on the lower wing. Two research papers (James C Moss , and later F.M. Catalano and G. L. Brand) concluded that adding a Gurney flap to the main (bottom) element of a dual-element wing added downforce and improved L/D ratio. By fiddling with the Gurney flap height, overlap, and gap, they increased lift by 12% and increased L/D ratio by 40%.

But before you go adding a Gurney flap to your double wing, you should know that the authors only got those results after tons of experimentation. The height of the Gurney flap, the distance (gap) between the wings, and the overlap between the wings all need to be set correctly to get the most out of it. Knowing all of this, if you’re going to put an upper element on a Wing Logic wing (or any wing with a Gurney flap), you’ll need to be able to adjust the upper wing’s X-Y-Z coordinates for angle, gap, and overlap.

If this is all too much work for you, go and buy a 9 Lives Racing Big Wang and add The Deuce double element kit. It’s already set up with the right overlap and gap, and is simple to adjust for angle. The performance is excellent, and you will not be disappointed. Tell Johnny I said hi.

But if you’re a DIY-or-die kind of person (ahem, guilty), or you have more time than money, then maybe putting together your own dual wing how you want to spend a day. If that’s the case, read on and I’ll walk you through how I made a dual element for a Wing Logic.

Assembling the upper wing

I make wings rather than buy them, mostly so that I can experiment with different shaped airfoils and construction methods. My S1223 is a torsion box, and my MSHD is a foam core with fiberglass. But neither of those construction methods works great for a wing with a much smaller chord and less thickness. So rather than build one from scratch, I bought a couple cheap extruded aluminum wings on Amazon for $35 each. You can sometimes find them cheaper, and my friend Bill Fischer of Garage Heroes in Training once bought one of these wings and got a box of 10 for the same price.

Cheap extruded wing from Amazon, eBay, etc.

I’m not exactly sure what the airfoil is, but it looks a bit like a Wortmann FX 72-MS-150A. With a cL of 1.8, this is decent, but not what I’d call an ultra-high lift wing. According to my Car Wing Comparisons article, the airfoil outperforms the NASCAR used for in their Car of Tomorrow for a hot second.

Airfoil Tools is a great place to research wings.

These cheap extruded aluminum wings are strong and light. They have two internal semi-circular spars that run the length of the wing, and provide a lot of stiffness. These supports are also tapped with M8 threads and do double duty fastening the end plates. While I might wish for a different shaped airfoil, the entire design is lightweight, sturdy, and inexpensive.

The wing has a 4.7” chord, which is larger than the upper element 9 Lives Racing uses. A rule of thumb is that the upper element should be about 30-40% the chord of the total wing (combined chord of main and second element), and this second element comes in at 32% of the combined 14.7”, and that’s right in the ballpark.

The longest of these cheapo wings I’ve found is 135cm (53.3”), and so if you want a bigger wing than that, you’re going to have to figure out a way to join them together. Welding is the obvious solution, but I didn’t want to rely on skin strength alone, I wanted to add an internal support as well.

I cut threads into one of the wing holes and installed a M8 stud, bottoming it out on the threads. Then I tapped the same hole on the other wing. I sandwiched a little bracket between them, which will be used to hold up the center of the wing, and then twisted them upon each other, essentially threading the two wings together.

I took the wing to a local fabrication shop and they charged me their hourly minimum of $80 to weld it up. So that’s $150 for the upper wing, all in. I’m sure the welding could be done cheaper, especially if I was doing several wings at the same time.

Welded all the way around, and pivoting on the center support.

Double wing end plates

To mount the upper wing to the lower, I’d need to make new larger end plates that hold the ends of the upper wing. The top wing also needs to be able to adjust for angle, gap, and overlap, and because it fits inside the end plate, it’s kind of an end plate within an end plate situation. I made the inner plates from 9mm plywood because I needed to countersink the 8mm hardware into the ends. If I used 12 gauge aluminum, the bolt heads would stick up proud and keep the wing from changing angle.

Maximum angle for a second element is typically around 40 degrees, measured from the bottom element. But at this angle, the upper wing risks flow separation. A safer bet is to set the upper wing to 35 degrees, which should provide nearly the same downforce as the maximum angle of attack. I traced all this out on the end plate (a No Parking street sign, per my usual $1-per-pound source at the metal recycler).

I always lay out the chord line parallel to the upper edge of the end plate, this makes it easy to set the angle of the wing. I’m also mocking up the position of the upper wing.

I first made the maximum downforce 35-degree setting, and to this I added a low-drag setting of 25 degrees. I don’t see needing any more adjustment than that, because I can always rake the entire wing to adjust between the high- and low-downforce settings. If I want less downforce, I’ll just remove the upper wing and run it as a single. From there I can tune wing angle and Gurney flap height as I would any other single element wing.

Upper wing pivots inside of end plate. You can see the forward hole, which increases overlap and gap. The gaps are larger than you’d have normally, because of the Gurney flap.

The completed double wing weighs 22.8 lbs total, and so the upper wing added only 6.2 lbs, including all of the things required to mount it. That’s pretty light, and it feels quite sturdy. Eventually I’ll lighten the main wing by milling out slots and wrapping it with carbon fiber. But more on that DIY project when I’ve liberated the wing from Steve.

Data?

This section is supposed to be filled with A/B testing data, including vital details about the ideal gap height for a dual-element wing that has a 1/4” Gurney flap on the bottom wing…. but instead it’s filled with a pissy rant.

Steve and I had a full test day planned, which involved him setting a few laps and then coming into the hot pits, where I could quickly change the main and upper wing angle, gap height, and swap between single vs dual wing. But despite an entire day at Watkins Glen, we got shit all of nothing. The problem is the same as the first time I did aero testing… Watkins Glen.

The weather is always variable, and the first session was wet and made data irrelevant. In the second session, a McLaren (620R?) dumped it’s coolant and oil on the first lap. This sent four cars into the T11 wall, and the cleanup crew onto the track for a lengthy stint. In the third session, again on the first flying lap, a Corvette stacked itself in Turn 2, requiring a full session of cleanup. And in the fourth and final run of the day, a BMW M2CS decided to get some new baby-blue racing stripes in T10. In the end, I don’t think the Advanced/Instructors run group got more than 15 minutes of track time the whole day.

Now this is the same run group I would have been in if I chose to drive that day. The two people I was with (Steve and Gregg) were the first two cars through the oil. Steve was going slowly because the McLaren directly in front was misting oil on his windshield. Gregg went through at speed and saved it like a hero. But he has a ton of experience at WGI and has proven many times over that he can save a spin.

Gregg saves it and avoids the wall. The next four cars don’t.

Well, if I was out there, I would have certainly been passing both of them in the session, which would have made me the first car through the oil. Dodged a bullet right there, I did! (I’m kidding about passing them; I drive like a grandma on this track.)

And this is why I seldom drive Watkins Glen, even for free. There are so many other tracks that have runoff, sand traps, and slower speeds, and are much safer as a result. Where I find enjoyment is pushing the car to the limit, and I’m not going to do that here, it just doesn’t make sense, financial or otherwise. My understanding is that some track day insurance companies will no longer cover cars at Watkins Glen, and I can’t blame them for that.

But I also understand that many of you like the combination of high speed and steel walls; you feel it gives you focus or commitment or whatever. Good for you. But the reason i have no data or wing gap information is because someone else also felt that way, and lost their focus or commitment or whatever.

Mounting a Wing Logic wing on a Subaru WRX STi

Wing Logic makes a satisfactory wing at a great price. It doesn’t come set up for any particular car, and so you will need to do some DIY fabrication. You’ll have to figure out things like how to mount the wing securely in the correct location, and then you’ll need to drill the bottom mounts for a range of useful angles, and then weld the bottom mounts to the wing.

It takes some patience and know-how to figure it all out, and if I’m being honest, 90% of people will cock it up and do a sub-optimal job, losing some of the wing’s performance in the process. If you buy a ready-made kit from 9 Lives Racing, AJ Hartman Aero, and other reputable companies, you’ll get an easier install and better performance straight out of the box.

But if you know what you’re doing, a DIY wing like Wing Logic or 9 Lives Racing’s new Express kits, can save you some money. Or if you have a car that doesn’t have a ready-made kit available, then a DIY solution is the only game in town.

I tested a Wing Logic wing in the wind tunnel vs the industry standard 9 Lives Racing Big Wang (Wing Logic versus 9 Lives Racing), and found the performance was similar. Wing Logic’s wing has less camber, and so it makes less downforce for a given area, but it’s a physically larger wing (9.2” vs 10”), and so the performance ends up being quite close. 9 Lives has the edge in weight and performance, and so if you’re racing in a series that limits the total wing area (GLTC, SCCA TT Nats, etc), or if you care about two pounds (high up, at the polar end of the car), then 9 Lives is a better choice. For the track rat or hard-park poseur, it’s a wash. I have plans to radically lighten the wing by milling it out into a skeleton, and then wrapping with carbon fiber, but that experiment will have to wait until the end of the track season, because I no longer have it.

Steve Leo is making a foam composite dual element, and while that process takes place (for fucking forever) I loaned him my 65” Wing Logic for his Subaru WRX STi. He didn’t have any wing mounts for it, but he had a spare trunk, and I said bring it over and I’ll figure something out.

Because I had already tested this wing on a Miata, the wing has brackets welded 41” apart. Ergo, I’d need to put the same spacing on the Subaru trunk. As luck would have it, that put the wing mounts right on top of the hinges, which is a sturdy area with some extra thickness and support in the metal frame. Also as luck would have it, the hardware would now be in the way, and I’d have to work out how to allow the trunk to close.

But first things first, to place the wings on the surface of the trunk, such that the wing is braced against side to side movement. I went to the Lowe’s racing department and bought a couple feet of 1” angle aluminum. I cut this into four pieces to make brackets, and sandwiched the Miata wing stands between them.

But I couldn’t put bolts through the top of the trunk and nuts on the bottom, as the nuts would be in the trunk gutter – this would cause interference and keep the trunk from closing. So I installed the hardware upside down, with countersunk 6mm bolts going through the underside and nylock nuts on top. It’s maybe a little less attractive, but this was the only way I got it to work.

Countersunk bolts of differing lengths install from the bottom.

It took a bit of head scratching, but I figured out how to install all 12 bolts from underneath. It took bolts of different lengths, which I’ve noted in the previous images in case you want to give it a go. It all came out surprisingly sturdy, and probably more rigid than you’d see on most cars. You can grab the wing and shake the car side to side, and the wing mounts don’t move.

Angle aluminum brackets on the trunk hold the wing stands in place. It’s very rigid.

The only thing I wasn’t terribly happy with is that the wing is too far forward. The rule of thumb is to overlap the trunk by 1/4 of the chord. So on this 10” wing, 2.5” should overlap the trunk, and 7.5” should be over the bumper.

How important is that? Well, I tested a Civic coupe in the wind tunnel, and moving the wing from on top of the trunk to the ideal rearward position resulted in less drag and more downforce. Usually you get one at the expense of the other, but this was a win-win. I would have liked to do that for Steve, but I didn’t have any more aluminum stock to make another set of wing stands, and he needed this done ASAP.

We got super lucky with the setback distance, any closer and the wing would hit the roof when the trunk was fully opened. I’ll take that as a win and optimize the wing stands at some later date.

By a stroke of luck, the trunk can open fully and the wing doesn’t hit anything.

Data?

Steve reports that the wing is working really well, but through some bad luck, we have had the devil of a time getting comparative A/B data. He’s used the wing at Watkins Glen, Lime Rock, NYST, and Pineview Run, and while we have data from those events, it’s not the kind of back-to-back data that tells a compelling story.

By the seat of the pants, the Wing Logic setup works better than the VSC rally wing he had on there previously. But the Wing Logic single wing doesn’t work quite as well as the Wing Logic dual element he tried this week. Wait, what?! Dual-fucking-element Wing Logic? You can read about that in the next post.

Wing-Logic vs 9 Lives Racing

Note: This article compares the original Wing-Logic airfoil versus the original 9 Lives Racing Big Wang. This article is still relevant for people with those older wings, but both wings have been updated. For a comparison using the newer Wing-Logic MSHD, see Wing-Logic MSHD: Wind Tunnel Tested.

The availability of inexpensive extruded aluminum wings has changed motorsports. Ten years ago it was rare to see a wing on a budget endurance racer or HPDE car. Now they are everywhere. A lot of the credit (blame?) goes to Johnny Cichowski of 9 Lives Racing, who sells an inexpensive extruded aluminum wing that sets the standard for the industry.

When you have a winning formula, people copy you. That’s the nature of business. That’s also the nature of racing. And so it’s no surprise that someone else has created a similar product.

This new extruded aluminum wing comes from Michael Jui of Wing Logic. He got wind that I was going to a wind tunnel, and asked if I would test his product. Of course I said yes, because I want that data, and I think the racing community as a whole would like that as well. So let’s take a look at this new wing in as much detail as I can, and compare it to the 9 Lives Racing Big Wang.

Wing Logic’s chord measures about 9.8”, while the 9LR wing is 9.2” across the top. So, the Big Wang is actually the smaller wing.

The wingspans are slightly different, as Wing Logic’s comes in standard lengths of 60″, 65”, 70″ and 72″. 9 Lives makes their wings to fit certain models, but you can also order a custom length. This is important because most racing rules limit wings to body width, and so if you have a Miata, you need to trim an inch off the 65″ Wing Logic, and re-tap the holes on one side. I didn’t do that for this test, so Wing Logic has a slight advantage, with 1″ more wingspan.

The construction of the wings are pretty similar, but Wing Logic has some extra internal thickness in a couple places, and while 9 Lives uses 5mm hardware for the end plates, Wing Logic uses 6mm. The Wing Logic end plates are 3mm, which is thicker than I’ve seen anywhere. All of this adds up to a wing that is sturdier, but heavier. A 65” Wing Logic fully set up with end plates and welded bottom mounts weighed 16.6 lbs. A 9 Lives 64” wing dressed the same weighed 14.6 lbs.

The shape of the wings are similar, but not the same. Anyone who says Wing Logic copied the 9 Lives Racing shape is simply wrong. As near as I can tell, Wing Logic is using a CH10 (Chuch Hollinger CH 10-48-13), which is a low Reynolds high-lift aviation airfoil, while the 9 Lives Racing wing was designed as a motorsports wing from the start.

If you measured Wing Logic in Benzing coordinates it’s a Be 133-105, while 9 Lives measures Be 123-125. That means the position of maximum thickness and maximum camber are the same in both wings, and that’s why they look similar. But Wing Logic’s is thicker and 9 Lives has more camber.

If you want to dive deep on that topic, I wrote an article on Car Wing Comparisons. Of all the wings I investigated, the most efficient was the CH10. But that is based on free stream efficiency, which means that the wing is not on a car, but just suspended in the air.

Cars are large and aerodynamically inefficient compared to wings, and when you put a wing on a car, the drag doesn’t go up that much, but the downforce does. So you usually get the best vehicle efficiency by choosing a wing that has the most downforce, not by choosing a wing for its efficiency or low drag.

Enough preamble, how do the wings compare in the wind tunnel?

Wind tunnel testing

I use the A2 wind tunnel, because for people like me, it’s the only game in town. I’ve tried to get into other wind tunnels, and either I’m not allowed, or it’s prohibitively expensive. The Aerodyne wind tunnel next door has a rolling floor, and I’d like to try it, even if it’s 4x the cost. But the rollers are designed for NASCAR, so unless I bring a car with a 109″ ish wheelbase (Miatas are 90″), I’m shit out of luck.

A2 doesn’t have a rolling floor, and so the effect of the tires rolling on the ground can’t be measured. This is significant for underbody testing, but you can measure things on the front of the car and over the body with acceptable accuracy.

There’s a lot of online conjecture by people who have never been to A2, that the small size of the wind tunnel, and the proximity of the walls, contributes highly to a blockage rate (or ratio?) that makes this wind tunnel results inaccurate. However, the walls in the tunnel are designed to reduce the blockage, by being curved rather than flat. And I’m also not after 100% accuracy, I’m looking for deltas – the difference between this or that. Did this wing have more or less drag than the previous wing? Did it produce more or less downforce? These things can be measured with accuracy at A2.

For the wind tunnel testing, I wanted to to use a car that already had CFD done on it, so I borrowed Phil Sproger’s car. His NA Miata has a 9 Lives Racing medium aero kit, and we set the ride height so that it would be as close as possible to the specifications that Morlind Engineering used when they ran CFD for 9 Lives Racing. (AJ Hartman and I did 42 runs on this car, baselining the car vs CFD, and then testing wings, spoilers, fastbacks, and many other ideas for drag and downforce. You’ll be able to read about that in a future report.)

We tested a 65″ Wing Logic back to back with a 64″ Nine Lives. Wing Logic has a built-in 1/4″ Gurney flap, and 9 Lives Racing has a slot for Gurney flaps of various heights, and we used a 1/2” Gurney. But because the slot is slightly recessed, this is more like 3/8”. So while one wing was slightly larger, the other had a slightly taller wicker, and it should be a fair fight.

Wing Logic – notice the smaller end plates.

I also threw in a strange DIY wing I made to see how it compares. It measures 41″x16″, and looks like an old-school F1 wing. The low-aspect ratio would not work in its favor, however the extra chord should make the wing slightly more efficient by having a higher Reynolds number.

Let’s see how they measured up at 100 mph, using the following fields:

Front downforce – This is a negative number, which you can think of as lift. When you add rear downforce (or drag), it lifts the front of the car through leverage, like a see-saw.
Rear downforce – This is the wing’s job, and varies with wing angle, Gurney flap height, and many other variables.
Total downforce – Front lift and rear downforce combined.
Drag – The number of pounds of drag the wing adds. This is a pretty meaningless number on its own, but is used to calculate the L/D ratio.
HP – An easier way of thinking about drag is how much horsepower it consumes.
L/D ratio – This is also known as aerodynamic efficiency, and is usually a good way of determining which part is better on the car. Note that the numbers in this table are not the bullshit free-stream efficiency you see in CFD, but the actual returned efficiency as run on the car.

Three wings compared at zero and 5 degrees.

As you can see from the data, at zero degrees angle of attack, Wing Logic, 9 Lives Racing, and my DIY wing are all better than a 10:1 L/D ratio. At this setting, the Big Wang makes the most downforce, and is thus the most efficient by about 5%.

When I set the wings to 5 degrees, Wing Logic and 9 Lives are identical in terms of aero efficiency. The Big Wang makes slightly more downforce, but the bigger wing makes less drag, and they are effectively the same at this setting. This is also about the maximum angle you can set, because air is coming down the roof at 5-7 degrees, and so if you use more angle than this, the center of the wing stalls, which adds drag and reduces downforce.

As a side note, my chunky DIY wing was similar to the other wings at zero degrees, by virtue of less drag. I should have tested it at the same 5 degrees as the others, because it was stalling at 7 degrees. It’s definitely the worst wing here, but not by a lot. On the plus side, it weighed on average 6 lbs less than the aluminum wings, and it cost me $30 in materials!

To get back to the actual contenders here, the two aluminum wings performed the same at 5 degrees AoA, but that isn’t the whole story, because the Big Wang is dimensionally the smallest. If you are racing in a class that limits total wing area (GLTC 500 square inch wings, for example), then the 9 Lives wing will give you about an 8% advantage over Wing Logic. It’s not a huge difference, but in a rule set that limits wing area , I’d take the Big Wang.

The full results of the wind tunnel testing are in my Miata Wind Tunnel Report, where I go into not only wings, but different tops, front end options (canards, hood and fender vents, splitter adornments, etc. You can purchase the report here by clicking the button, and you’ll get a link where you can download it.

Purchase Miata Wind Tunnel Report

If you don’t want to invest in the wind tunnel report, but you want to support this kind of investigative content, Buy Me a Coffee. It takes 15 coffees ($5 each) to pay for one run in the wind tunnel. That doesn’t count the many hours of preparation, towing 11 hours each way, gas, hotel, and other expenses I accrue on the way. Your support makes it possible for me to do community-driven wind tunnel testing. Thanks!

CFD vs wind tunnel

Both 9 Lives Racing and Wing Logic publish CFD results of their wings, and I thought it would be illuminating to see how accurate those are compared to the A2 wind tunnel. Wing Logic use Kyle Forster to do their CFD, and he used a 70” wing on a Mustang rather than a 65” wing on a Miata, so it’s not a direct comparison. Kyle used the same CFD settings as he did for AJ’s wings, which have generally landed within about 3% on loads in the same tunnel with matching cars/positions etc, so A2 correlation should be OK for this data.

Note that the reference area for the Miata is 1.67 meters squared (18 square feet), while in the CFD, it’s set at 1 meter squared. Ergo, I need to multiply the CFD numbers by 1.67 so that they match the wind tunnel data. When I do that, you can see, the wind tunnel returns similar downforce as the CFD, but drag is higher in the simulation.

Wing Logic wind tunnel (65″ Miata) vs CFD (70″ Mustang)

Is this a big problem? Not really. After testing the same wing on a Veloster N, an 8th Gen Civic coupe, a Miata, and a Supra, I can tell you that the shape of the car has a greater impact on a wing’s performance than any CFD error. On a hatchback, I’ve seen a 3.5:1 L/D and on a coupe up to 24.5:1. In the end, the main difference between CFD and wind tunnel results isn’t computer vs real world, it’s the shape of the car. A Miata ain’t a Mustang, and that’s likely where most of the differences lie between the Wing Logic CFD and wind tunnel data.

9 Lives Racing also publishes CFD using a 64″ wing on a Miata, and that’s exactly what we brought to the tunnel. So we can get an apples to apples comparison on this wing for shiz. But 9 Lives doesn’t publish coefficients, they only give us drag and downforce. Still, it’s enough to work from.

9 Lives Racing wind tunnel vs CFD at 100 mph.

Comparing the wind tunnel results to CFD, the computer predicted both less drag and less downforce than we got in the wind tunnel. The fact that the CFD was off by 33% in downforce points to some kind of problem with that model.

I base that on another nugget Kyle dropped on me which is that as a general rule of thumb, if CFD is off by <20% it could be correlation, <30% it’s setup differences, >30% something has gone wrong with the output numbers at some point or the CFD is extremely broken. So I’m not sure what’s going on with the 9 Lives / Morlind Engineering CFD, but it’s not matching real-world values very closely. When we tested the whole car (not just the wing), the drag was also off by a lot (as in .4 vs .6).

70″ Wing Logic, end plates, 3/4″ Gurney

I was also able to test a 70″ Wing Logic on my Hyundai Veloster N, which allowed me to compare the effect of wingspan on a hatchback. Last year I tested a 55″ Big Wang on my Veloster, and since Wing Logic and 9 Lives are pretty similar, I can get find out how much I was giving up with a shorter wingspan.

I also wanted to try 12″ square end plates to see if the smaller end plates were giving up some performance. And since Wing Logic has a built-in Gurney flap, I also wanted to see what would happen if I butted up a 3/4″ Gurney flap against that. (If you’re wondering what a 1/2″ Gurney would do, you can just average the numbers).

Comparing the 70″ Wing Logic to 55″ 9 Lives is absolutely not fair, so think of this only as a comparison of wingspan. The longer wing made 10 lbs less downforce (what in the actual fuck?), but had a lot less drag. The end result is a much better L/D ratio of 7.4:1 for the longer wing compared to about 3:1 for the shorter wing.

Now this is flatly absurd, because according to Kyle, an increase of 100mm span is worth about 7% load on the same car, so an increase in 15″ span (375 mm) should be an increase of 26% difference in load, not a decrease of almost 7%. I’m not sure what’s going on here, as the wing was in a similar position, and on the same car.

Changing the smaller end plates for 12″ square end plates added 7.2 lbs of downforce, but also used an additional 1.1 hp in drag. The change in parts returns a L/D ratio of 1.3:1, which means the smaller end plates would be faster anywhere but an autocross course.

Adding a 3/4″ Gurney flap added 48.1 lbs of downforce for 21.4 lbs of drag, which works out to a 2.25 L/D ratio. If you need more rear downforce, adding the larger wicker would be worthwhile on just about any track that isn’t a high speed oval.

Conclusions

Comparing the two wings, my main gripe is that Wing Logic’s wing is 2 pounds heavier than a 9 Lives wing of the same length. That’s not a lot of weight, but the location of the weight, high up and at the polar end of the car, isn’t helpful. As previously stated, in a racing class that limits the wing size to a certain amount of total area, 9 Lives has a slight advantage.

Those negatives aside, Wing Logic has a fine product that is well constructed and performs similarly to the Big Wang. At $349-399 with free shipping, Wing Logic’s wing is a downright bargain.

Testing Wing End Plates in a Wind Tunnel

The following article is made up of excepts from my wind tunnel report. You’ll get a more cohesive story, and a lot more data on many more aerodynamic parts, if you buy the report and read it end to end.

End plates on wings are necessary; they separate the low-pressure region under the wing from the high-pressure area on top of the wing. The suction side of the wing is what does most of the work, so by keeping the high pressure side from bleeding into the suction side, the wing makes more downforce.

The shape of the low-pressure region under the wing is different for every airfoil. However, for most wings designed for motorsports, you’ll find that the low-pressure region is at the front of the wing and often extends ahead of the wing. The low-pressure area extends about a chord’s distance below the wing as well.

*The shape of the low-pressure region below the wing depends on the airfoil. I inverted the images so that it relates to car wings. This image is from Race Car Aerodynamics, and I highly suggest you* buy the book.

Given that information, and after looking at the preceding image, you might conclude that a good endplate should be shaped to exactly cover the high and low pressure regions of the wing. And from that, you might surmise that a good endplate for a 10” chord motorsports wing should extend 10” below the wing, and should have a lot of surface area concentrated at the front. And that’s how I see it as well. However, some end plates have most of the area towards the rear of the wing, and I can’t say I understand that. But aerodynamics is full of weird contradictions, and perhaps some of those end plates work.

With the amount of companies selling improved end plates with different shapes and sizes, you’d assume there was something to be gained over a plain rectangular end plate. And because some of these fancy end plates cost a couple hundred dollars, and boast CFD-designed pedigrees, they must be doing something useful, right?

CFD and wing efficiency

Some of my wind tunnel data conflicts with published CFD (computational fluid dynamics) data. This isn’t surprising, as CFD is just a computer calculation, and not real-world data. Manufacturers typically test wings in free-stream CFD, meaning that the wing is suspended in mid-air, as only a computer simulation can do. This is the best way to calculate what happens when you change wing angle, add Gurney flaps, or change the shape and size of end plates. Free stream CFD is essential, because it eliminates everything in front of the wing. This is really the only way to compare one thing to another.

But when you put a wing on a car, everything in front of the wing affects its performance. Wind speed and direction, cars in front of you, and open windows can make a huge difference. Plus there’s the shape of your car, the angle of the windshield, aerodynamic devices on your car, like splitter, canards, hood vents, vortex generators, GPS antenna, wing stands, … you name it, every single thing that’s in front of your wing changes how it performs. You’ll never get the same amount of downforce from your wing as the free stream CFD data shows. Not even close.

You can research wings on Airfoil Tools or Bigfoil, or use tools like Javafoil and CFD, and you’ll find wings that have a 14:1 L/D ratio, or better. But when you put the wing on the car and adjust the angle of attack, you’ll be stoked when your wing has half of that.

For this reason, anything you do to improve the efficiency of the wing in free-stream CFD is meaningless until you put it on the car. For example, modifications to wing end plates can reduce drag, and this shows up in CFD as a gain in wing efficiency. But when you put the wing on the car, the drag of the wing is inconsequential to the total vehicle drag. Touring cars are essentially huge rounded bricks, and wings are tiny streamlined objects by comparison, and so you can understand that the drag from the wing is essentially nothing compared to the drag of the vehicle.

In reality, the only thing that matters is the aerodynamic efficiency of the entire vehicle, and you typically get that by going after as much wing downforce as possible. Modifications to the end plate that reduce drag might increase free stream wing efficiency, but they do that by reducing wing downforce. And this makes the L/D ratio of the car worse, and the car goes slower. Ergo, it’s utterly worthless to optimize wing efficiency in free stream CFD by reducing drag. If you use CFD for anything, it should be for optimizing the wing for maximum downforce.

Testing end plates in a wind tunnel

Before I get to the testing data, let me tell you exactly how shit stupid I am. My aerodynamics sensei Kyle Forster had these things to say about end plates:

Use a rectangular shape. Adding vents, cuts, and other tricks are more likely to reduce the performance of the wing than improve it.
Optimizing the performance of the wing end plates is the least important part of the entire vehicle’s aero package.

You might think I would take Kyle at his word. He worked for the Mercedes Formula 1 team as an aerodynamics engineer during the manufacturer’s most dominant years. But I’m also a stubborn, pig-headed ass who believes in getting his own data. So I took three end plates (four if you count that I turned one backwards) to the A2 wind tunnel and spent my hard-earned money to see if he was right.

I tested four wings in the wind tunnel, but when I got around to testing the end plates, they were all swapped onto a 55″ (1397 mm) 9 Lives Racing wing. This is the benchmark motorsports wing for many good reasons, so I figured why not go with the industry standard.

I first tested the basic rectangular end plate, which is made from an aluminum street sign that I simply cut in half, rounded the corners, and called it done.

Sorry about the image quality, these are stills from the video monitors in the wind tunnel.

I then swapped those for a popular CFD-designed end plate. I’ve always found this design to fly in the face of reason – why is there a big cut out right where the low-pressure region is?

CFD end plate with a pressure relief cut on the top, and a large radius cut into the leading edge.

I then turned the end plate backwards and tested that. From my point of view, it seems like the end plate might perform better with more surface area facing forward and less at the rear. It wasn’t a great fit, though.

The CFD end plate didn’t fit very well when I flipped it around backwards. I got two of the holes to match up and called it good enough.

Finally I tested an end plate of my own design. It’s in some ways the opposite of the CFD end plate, having a lot of surface area forward and tapering towards the rear. There’s a very small relief cut on the upper back corner that’s supposed to reduce a vortex there (er… so I’ve read). But more significantly, this end plate has a very small wicker on the trailing edge.

My Occam’s Racer end plate with more area forward and Gurney flap. This is the same end plate in the cover image.

Let’s see how the end plates performed:

End plate	cD	cL	Vehicle L/D
Rectangular	.467	-.382	.82
CFD	.475	-.386	.81
CFD backwards	.474	-.385	.81
Occam’s Racer	.480	-.398	.83

Coefficients of drag and lift with various end plates. cL is a negative number because it’s showing downforce; more negative is more better.

Wind tunnel data

So let’s unpack the coefficient data and translate that into more common figures, like pounds of downforce and horsepower consumed.

Rectangular – Vehicle L/D ratio .82
- Baseline to compare with other end plates
CFD – Vehicle L/D ratio .81
- +1.8 lbs total downforce
- 4.1 lbs drag = 0.4:1 L/D ratio for end plate
- +1.1 hp used from drag
CFD backwards – Vehicle L/D ratio .81
- +1.5 lbs total downforce
- 3.5 lbs drag = 0.4:1 L/D ratio for end plate
- +.9 hp used from drag
Occam’s Racer – Vehicle L/D ratio .83
- +8.8 lbs total downforce
- 6.9 lbs drag = 1.3:1 L/D ratio for end plate
- +1.8 hp used from drag

The first thing you’ll notice is that the CFD end plate increased downforce by almost 2 lbs at 100 mph. That’s more than the rectangular end plate, but not much. As a consequence of that additional downforce, there’s a bit more drag.

Turning the CFD end plate backwards resulted in less downforce, but also less drag. Overall, the CFD end plate performed the same forwards as backwards. Surprising.

Another surprise was that my Occam’s Racer end plates gained 8.8 lbs of downforce over the rectangular plate. Not surprising, this also resulted in more drag. However, these end plates resulted in the best vehicle L/D ratio.

Racing simulations

Those numbers are all very close, and you might be wondering how they affect the only thing that matters: lap times!

To find out, I put the coefficient of lift and drag values into OptimumLap and ran them around two race tracks, the autocross course from 2010 SCCA Solo Nationals, and Lime Rock Park. I typically use these two tracks because they are close in lap time, but are completely different with respect to speed. I’ve also included my local track Watkins Glen, because it’s very high speed and should spread the results out more. (My wind tunnel report shows lap time comparisons for every part that I test, as well as some useful combinations.)

End plate	cD	cL	Autocross	Lime Rock	WGI
Rectangular	.467	.382	61.79	61.03	134.74
CFD	.475	.386	61.79	61.05	134.81
CFD backwards	.474	.385	61.79	61.05	134.81
Occam’s Racer	.480	.398	61.78	61.04	134.81

Lap times

On the autocross track, my end plates won by a whopping .01 seconds. At Lime Rock, the rectangular street signs won by the same insignificant margin. At Watkins Glen, the “No Skateboarding” sign went .07 seconds faster than either of the fancy end plates.

Discussion

The CFD-designed end plates were a disappointment, and put an exclamation point on my rant at the beginning of this post about using free stream CFD. Look, I’m not at all doubting that these end plates worked better in CFD and returned exactly what they calculated. But overall performance didn’t change facing either way, and shows how useless free stream CFD can be in the real world.

The custom end plates I designed are based on a hunch that I should put most of the area low and forward. More significantly, these end plates have a small Gurney flap on the outer edge, and it’s likely that the shape of the end plate was no better than the others, and it was simply the addition of the wicker that gave this end plate the most downforce. Most downforce doesn’t mean best, as it wasn’t terribly efficient and only returned a 1.3:1 L/D ratio above and beyond the rectangular plate. Using these end plates would make the car faster on a tight track, but slower on most race tracks.

In the end, there really is nothing wrong with a rectangular end plate. If you don’t have the means of CFD testing your entire vehicle and optimizing the end plates to your entire car, then a rectangular end plate is your best bet. I make mine from street signs that I buy from my local metal recycler for $1 per pound. And so both end plates are less than $2 and come with amusing graphics. It astonishes me that people will pay much more for something that performs worse.

But let’s face it, no matter what you do to the end plate, it’s not going to make it significantly worse, either. End plates with cuts and vents are a great place for personalization, they are a source of many silly conversations, and they throw competitors off the scent. In the end, I say do whatever you want, it won’t matter much anyway.

But if you want to hedge your bets and do the least work, just stick to a rectangular end plate. Learning this lesson cost me $300 plus a lot of time and effort, and it proves I should just have just believed Kyle Forster.

But it wasn’t a total waste of time and money if that keeps other people from making the same mistake. So if this article saves you money, please consider buying me a coffee, or if you want 50+ pages of the same kind of data, then buy my wind tunnel report. It’s only through contributions that I can afford to do wind tunnel testing and continue the lord’s work. Thanks!

Postscript

After posting this article, I got a bunch of great comments on the Professional Awesome Technical Forum on Facebook. This group has a wealth of knowledge that surpasses my own, so please see this post, scroll through the comments, and benefit from the global knowledge base.