How Spoilers Work

Buckle up, this one's long.

After hanging around Ecomodder for years, I had a pretty messed up idea of the function of spoilers because of what I had read on that forum. You might too.
 
See, the prevailing theory there—based entirely on posts by one forum member who is widely seen as an aerodynamic “guru”—is that spoilers “reach out” to an ideal “template” profile. By so doing, the theory goes on, a “locked vortex” is captured and the airflow above the spoiler follows the “ideal” template line as the air just ahead of the spoiler recirculates. So, simply position a spoiler and extend it until it touches a template profile overlaid on a side-view image of your car and, voilà! Lower drag—because the spoiler isolates negative pressure in front, higher pressure in the wake behind it, and the outer flow follows the “template” shape.

Someone might want to tell this person that the "perfect match" RS spoiler was, according to Burst himself, supposed to be 15 to 20 mm longer.

The only problem with the theory is this: it is completely, utterly wrong. It depends on another assumption that has been promulgated on the forum, that merely intersecting a “template” somehow causes air to flow in the shape of that template. This untruth results in posts such as this:

"AST" here is shorthand for "aerodynamic streamlining template."

Yes, there are people out there who believe that the Murciélago:

(Image source: Autogespot).

…and Volkswagen XL1:

I took this one myself. You can see this car in person at the Lane Motor Museum in Nashville, TN (I highly recommend a visit).

…and Audi A8L:

(Image source: Wikimedia Commons).

…all have the same rear-end shape as far as airflow is concerned. This belief is so batshit crazy it isn’t even worthy of refuting. If you believe this, get off the internet and go read some books by actual engineers.
 
 An Ounce of Testing
 
What bugs me so much when I read posts by the originator of this theory (or others parroting him) is the fact that some simple testing on your own car is enough to immediately disprove it (in fact, some of the first testing I ever did was aimed at investigating this claim). Take my car for example. Here it is with a “template” overlaid:

I've written before about how this "analysis" resulted in a kerfuffle over the Prius' supposedly-separated flow over its rear window.

The red line indicates the height of a vertical spoiler to intersect the “template” profile, which—according to Ecomodders—will reduce drag by capturing lower pressure ahead of it and higher pressure behind. It’s about 5.5 inches tall, or a board angled from the base of the rear window at about 42 degrees.
 
Well, as it happens, I’ve been testing spoilers of various heights and shapes over the past few months. What did these spoilers do to pressures ahead of them (on the window) and behind them (on the base)?

This is identical to the spoiler I ran for the past two years—minus the garish faux carbon fiber finish.

I tested this one at three different angles, with and without fins.

I even bought a cheap Hellcat replica. Looks cool, but I probably won't fit it permanently without further testing.

Answer: exactly the opposite of what the theory predicts. Here's a chart of all the spoilers I've tested so far and how they change pressures on the window and base of my car:

Note that these are changes in pressure from standard configuration (no spoiler), not gauge pressure as I've sometimes plotted in previous posts.

Eight different spoilers, of varying height and shape--every single one increased pressure on the window, and all but one decreased pressure on the base.

I also tuft tested the spoilers while I had them in place. Here’s the Hellcat spoiler:

Flow separates in front of the spoiler if it’s large enough but the air there has higher pressure, verified by measuring pressures on the window. And the smaller spoilers on my car show no flow separation:

What’s going on here is different and more complicated than Internet Theory suggests, and reducing drag and lift by fitting a spoiler requires more thought and testing than simply “reaching out” to an imaginary template and calling it good.
 
What’s Really Happening?
 
Now that we’ve established how spoilers don’t work, that leaves the question of how they actually do work. This is a bit thorny since it's something that's still evolving, and you'll see slightly different explanations in older references than newer.
 
Here’s an excerpt from Aerodynamics of Road Vehicles, 4^th ed. (1998):
 
“By deflecting the flap (simulating the spoiler), the pressure on the flat plate (simulating the slant) is increased. If this modified pressure distribution is integrated in the x and y direction, the result is lower drag and lift. The isobars plotted in Fig. 4.85, measured for a fastback, confirm this observation. The spoiler causes a clear rise in pressure on the rear slope in front of it” (191).
 
And here is the same section on rear spoilers from the 5^th ed. (2016):
 
“As can be seen, by deflecting the flap simulating the spoiler, the pressure on the top of the plate is increased.
 
The isobars on the rear slant of a coupe plotted in Figure 4.112 seem to confirm the above analogy. The static pressure on the rear slant is significantly higher with the mounted spoiler than without it. Plotting the pressure coefficient c_p against the vehicle height z/h—as shown in Figure 4.112 on the right for the center section—the reduction in drag becomes evident. While the pressure on the rear of the vehicle is increased by the spoiler, it remains unaffected on the front side. Integrating the changed pressure distribution in the x- and z- directions results in decreased values for drag and lift.
 
However, a closer examination of the pressure distribution on the rear slant reveals a completely different working principle of this spoiler: The pattern of the isobars for the vehicle without a spoiler shows that familiar vortex pair, originating from the C-pillars, has formed. It manifests itself by high negative pressures near the lateral edges. The spoiler causes the vortices to disappear, resulting in a higher static pressure on the rear slant” (308-309).
 
The takeaway here is: rear spoilers increase pressure on the bodywork ahead of them by “spoiling” the airflow. You can easily measure this for yourself on your own car like I have and see what different spoiler sizes and angles do to the pressure distribution. Spoilers may also weaken the longitudinal vortices that shed off cars with fastbacks or notchbacks. These vortices are associated with increased drag and lift, and weakening them benefits both. Regardless of theoretical explanations, the spoiler works by increasing pressure on the body surface ahead of it--exactly the opposite of the "locked vortex" theory, which you can verify by testing is bunk.

Could a spoiler also help reduce drag or lift on a truck? I'll find out in coming weeks.

Perhaps the clearest, most succinct description of what spoilers can do is this passage in Julian Edgar’s Modifying the Aerodynamics of Your Road Car (2018):
 
“Rear spoilers are fitted so that rear lift forces are decreased, or more rarely, downforce is generated. They change the way that rear forces act in two ways. First, a flat plate angled to the airflow that causes air to be directed at any upward angle, will create a downward force. To be able to achieve this, it is best if the spoiler is placed in an area of attached flow, but such is the power of a spoiler, it will still have some effect even if it is working in an area of flow that is only partially attached. Second, and more importantly because of the area over which it acts, a rear spoiler can change the pressure that is being applied to the rear surfaces of a car. For example, a spoiler fitted to the trailing edge of a boot lid can cause an increase in pressure across the lid and rear glass. As mentioned, even a small change in pressure acting over a large area can be significant” (194-195).
 
You don’t have to take any of this on faith, contrary to what the internet theorists will tell you. Go test and find out for yourself.