This topic comes up regularly. A good thread is here:
Vortex Generators / Hmmm Airtabs

Probably it comes up because nobody has properly documented results, so it remains open for speculation.

The closest thing to experimental evidence I've seen for the application I'd consider (around the trailing edge of the car) is in the pressure tests illustrated in this post:

https://www.gassavers.org/showthread....dson#post22178

I'm not buying anything spelled like that. :-)

From the look of the images above and a few more aircraft ones I saw online, the VGs were always at the beginning, unlike the car image on the Vortekz site where it's at the very end. I have a fairly boxy wagon, and would probably try this out sometime. I should really find out if my freakin air dam does anything before I test something new though :D

Nah, don't let testing get in the way of trying other stuff... :-)

I need to finish the boat tail and test that. I really want to just skip the air dam test and just finish the boat tail. It's got to have a much larger effect.

https://www.gassavers.org/attachment....1&d=1185424999

Just look at that beautiful lack of seperation I expect! Top car is original image from a study, middle car is the shape of my car, bottom car is how I hope the air flows around my boat tail. I made those spiffy images in MSPaint too!

theclencher -

Ooooooh, here's another observation/opinion. For the Evo, there seems to be an assumption that the VGs work *with* the monster Evo spoiler. This design rule would apply to coupe/sedans, not wagons.

CarloSW2

i think cfg83's right
there seems to be a lot of misconseptions on what VG's do and where they should be placed.

if we look at airplanes where they are frequently used we can conclude they work. but al they do is delay laminar flow sepparation. at the end of a vehicle there's going to be flow sepparation anyway. their funtion to have more air follow the path of the rear window, and by the spoiler.

i think on a car they might also work when attached at the top of the front window to pervent a sepparation bubble to form, on older cars especially the angle from the windshield and the root was to steep for the air to follow. althought is would make a very impractical placement of the vg's

i think vg's could work on any car is they're placed and designed correctly. but this is difficuly without any propper airflow visualisation.

Yes, as a matter of fact, I have these pictures.
https://aycu38.webshots.com/image/242...2821607_rs.jpg

https://aycu06.webshots.com/image/233...9827658_rs.jpg
Dimensions are 1" high x 1" long
https://aycu33.webshots.com/image/244...4242620_rs.jpg

So, as you can see, they are modeled after VGs like the ones in the image you posted.

And also, I think this butter tray is compatible with all makes and models of stick-type butter and margarines :D

Aha. So one could also use extruded aluminum angle stock. Probably more work though. And maybe not compatible with margarine.

More seriously, for the rear edges of a wagon, do our aero geeks think these are better than a small spoiler - that is, better than an extension of the roofline or the side panel that would extend rearward a few inches? I'd consider building one that angles downward somewhat for a mini Kamm effect. Similar spoilers on the side edges could angle inwards, same concept.

Or maybe vortex generators instead? Or vg's in addition to a slightly angled spoiler?

Honestly, though you probably already know this, a boxy wagon is pretty much aerodynamically hopeless. There is no substitute for a boat tail on a wagon but VGs, in theory, are better than a small extension (small as in 6-8"). VGs create an aerodynamic simulation of a boat tail by energizing air flow, which delays the inevitable forming of turbulence and a resulting low-pressure area. Of course, unless you have them framing your roof and side panels they won't do much good. Ideally to compliment your roof and side panel VGs a smooth body pan and floor-mounted VGs would probably help a great deal.

Of course, you could use both the extension and the VGs, which would be even better than just one, as it would provide a little more space laminar flow and probably aid in the simulated flow attachment.

Here's a rough MS B&W paint illustration of what I'm talking about.
https://aycu16.webshots.com/image/230...2626130_rs.jpg