tuft testing on 93 Sentra back window
 

I was doing quite a bit of tuft testing this last couple of weeks. The back window is an ideal area for experimenting in general, because I can use the rearview mirror to get a real time look at what is happening.

It is very interesting. The flow along the top four inches of the window is very consistent, moving almost directly outward both ways from the centerline, then shifting to directly downward at the outermost three inches of the window. It is turbulent and not consistent enough to really comment about through most of the middle, with one exception. In the center of the window, about 6 inches down the glass, the tuft was always standing straight out from the glass.

If I had to guess what the airflow was doing, I'd say it was a pair of counterrotating vortices, flowing up the centerline, outward, and down the outside edges.

None of the locations of the little VG's that I built seemed to have any visible effect on the airflow down the back window, except for a slightly larger area at the outside edges where the flow shifted downward rather than being visibly turbulent.

The angle of the rear window is about 40 degrees down from horizontal, which is pretty bad. I've been wondering about a trailing 'hood' sort of thing attached at the top of the window, which would taper down at a more gentle angle, maybe 15 degrees or so, and let the airflow move a little more gently into the wake of the roof.

But, for this week, the only change I make before the next long highway trip is to remove the huge front mudflaps, and see if I get any noticeable difference. I've thought about using the rear mudflaps as an anchor point for a trailing aero dam, to smooth the airflow up from the bottom of the body into the rear bumper area. I want to do one thing at a time, so I can get a feel for which mods are the most valuable.

A friend gave me the old video camera and recorder out of a police car, and it might end up being mounted to watch tufts on top of the car and in places I can't see directly.

So many ideas, so little money... :D

I've been waiting for someone with a "3 box" car to try this, and yours sounds like a perfect candidate. You should simply mimic at the trailing edge of your roof/sides of the greenhouse what andrewj and I did on our hatchback cars:

See Post #78 in this thread:
http://ecomodder.com/forum/showthrea...amics-312.html

http://i239.photobucket.com/albums/f...rdkammback.jpg


http://metrompg.com/posts/photos/aer...ail-rear-s.jpg

It was, in fact, that car that inspired me to think in that direction. :D I have a hunk of white thermoplastic that may be OK to make a prototype. I think that if I do a permanent one it will be out of acrylic or polycarbonate. (Those are both thermoplastic, aren't they?) A smoke colored one could have some good effects visually as well as aerodynamically.

My goal is a reliable 50+mpg on the highway, which should be attainable.

I can't offer anything useful about thermoplastic. Cardboard is free, cheap and easy to work with for making prototypes. The only problem is rain.

I have more of a 3 smartie car but have been thinking of doing the same thing, building a Kammback to connect the top of the rear window to the trailing edge of the trunk. I sketched up a picture and and I get an angle of 16 degrees which from what I read is too steep to keep the flow attached.
I guess the next option is to try a kammback similar to the ones shown on the civic and metro hatch back. I can't imagine it looking very good but I'll try to sketch something up.
Ian
ps, this isn't my Neon, just one I found on the net. Mine is so much nicer with matching wheel covers.... ;)

3 smartie car - I like it. :)

A good rule of thumb is that you can begin with a 10 degree angle, and if your transition is gentle enough, that angle/tangent (if a radius) can be increased beyond that and still retain attached flow.

You could do worse than simply look at the rooflines of existing relatively aero vehicles, eg:

http://metrompg.com/posts/photos/prius-insight-a2.jpg

(from Kardboard Kammback: testing a partial boat tail prototype )

Also, don't forget it's a 3D shape - the transition from "side" to "top" is important, and should be generously radiused to aid pressure recovery & discourage vortex formation. If you look at my kardboard prototype, you'll see I used too sharp an angle at that transition point. AndrewJ did much better with his finished work.