I detect a note a sarcasm, but that's okay because I'm not completely sold on this myself. That's one reason not to spend money on it.

The best humor does have a touch of truth. While a tennis ball does have good turbulence the shape would be to big a drag. Reading your reply I remembered I have a roll of fuzzy material called mull skin. I was going to paint them but maybe the fuzzy stuff might work better.

https://www.carspace.com/hateful/Albu...o/v./photo.jpg

I attached three quickly thrown together VG's to the roof with a small piece of tape. The base is made from magnetic labels for warehouse shelves, so the little bit of tape held them on the roof at 55mph.

I attached some flagging tape to the rear window. The result was that the top two behaved about the same. The bottom two on the drives side behind the VG's layed down other that from cross wind at slow speeds, while the other two bottom pieces not behind the VGs flip-a-dee-flapped all over the place and I was expecting them to be torn off.

Apparently they do have an effect on air flow; how that equates to gas mileage I'm not sure.

Here's a pic from the Vortekz website. Its basically a modified graphic from the Mitsu PDF:
https://www.vortekz.com/images/withwithoutsm.jpg

The rear spoiler is kind of irrelevant, but notice how the low pressure area behind the car & rear glass is significantly smaller. That's where your potential gains would be.

that looks like what's going on. That graphic looks like one of those in the sticky. I'm convinced they do at least get drag off the back glass. I'm going to play around with these some more. The Cavalier starts with cd of .36, so I could yield some mileage out of it.

Cool.

The back of the car has separated airflow caused by low energy in the boundary layer. This bubble of separated airflow causes the strips of material that you put on the back of your car to flap around.

The vortices created by the VG add energy to the airflow and help reestablish the boundary layer and therefore reattach the airflow. Why is this a big deal? Because the bubble on the back of the car (before you added the VGs) acts as part of the body of the car, which in turn creates more drag than the "real" body of the car.

There is a famous experiment that most all aero weenies do in college, where the student measures the pressure at various locations on a sphere in a wind tunnel. The student then adds a wire "trip" to the sphere, remeasures the pressure distribution, and finds that the total drag on the sphere has been reduced. The "trip" acts similarly as VGs by reenergizing the boundary layer and "cleaning up" the airflow on the sphere.

Too much sciency stuff....neat pictures here:

https://www.princeton.edu/~asmits/Bicycle_web/blunt.html

(a nice writup on the ecomodder site as well)

https://www.carspace.com/hateful/Albu...o/v./photo.jpg

I made these from wood. I think the shallow holes will provide a little extra turbulence.

VG's the best way to reduce the wake...
 

I use the shark fin type the most swirl with the least amount of drag... I do say that placement on the cars roof is very important to proper flow... I had to change my placement a few times using a crude testing device but I was amazed on how well it really worked, this was done and shown max results...

I will try these on my civic:

https://www.buyairtab.com/

https://autospeed.com/A_3060/cms/article.html

"For our money, Airtabs are the vortex generators to buy. The company producing them makes claims which are more in keeping with technically feasible outcomes, the devices themselves are arguably better looking than folded aluminium designs (and certainly are much less dangerous protrusions to have mounted on vehicles) and are cheaper."

i like the plastic spoons :D

Keep us posted....I'd be most interested in your results.

BTW, is your Civic a sedan? Its difficult to tell from your garage photos.