you might remember the simpsons episode where homer is tolled by a salesperson the bullet holes in the car are actually "speedholes". here's why i bring it up:
i'm planning to try some more vortex generators etc on my car to keep the air from sepparating at the roofline. aircraft are the easyest place for inspiration as appart from the Mitsubishi Lancer and some trucks there's very few automotive examples.
some gliders however use a different method to cure flow separation... simply a series of tiny holes trough the wing, venting high pressure air from the bottom to the top.
so to put it bluntly, what would happen if someone would drill a series of small holes along your rear roofline or use a more sivilised method to vent interior air in small amounts into the boundry layer of the roof... the seem from the trunklid is conveniently placed on my car so perhaps incerting tubes between the rubber sealing might do the trick.also for example the vortex generators on the lancer could also include this system in an attractive way, adding nozzles to direct the air. i know there's been some experimentation by car manifacturers with air jet systems so it might work.... although passive pressure might not be enough?
could this have any benefits or are the pressure differences not big enough? also interior ventilation might improve.
As for interior venting, though I couldn't find any photos, see if you can find one of the area between the bottom of the rear window and the trunklid of a 1970 Chevrolet Impala or Caprice. Those vents were part of the system to evacuate the interior air. I have no idea what they did for aerodynamics.
after some experimenting last night i found that it should be possible to insert something between the trunklid and the rubber sealing. it will need to be rather strong, but i've been thinging about finding some kind of squared metal tube or something similar, for now i've inserted some plastic at one side. the sealing should still provide enough of a treshold to pervent water from seaping in... also the idea is air is sucked out rather than in so that should further prevent water from comming in while driving.
fortunately it's not raining now, but perhaps i can replicate this observation by sraying water on the window. or i could resort to classic tuft testing although that requires a but more time and effort. yet i could use this on the inside as well!
by keeping the initial setup assymetric any big changes in airflow should become visible. the air will be vented out trough the trunklid seem so there's a risk this will defuse the effect, but that remains to be seen.
Can you post a picture of what you are talking about ? ( A pencil sketch ? )
Also, this is one what kind of car ?
It sounds as if you are telling us that you are having some seperation at the back window of the car. You are trying to reattach the flow by creating a vacuum using negative pressure from other parts of the car ? ( Did I get that part right ? )
here's a picture of the rear window (the roof cones are no longer there,... previous experiment)
my car is an opel kadett hatchback... you can see it in side view in my garage. the rear window is part of the rear hatch so there's a seem above the window... the basic idea is to duct air from the interior to the outside, since i assume the pressure difference would be able to do this... since i'm not homer simpson i'm not going to actually drill holes in my roof, althoug this would accomplish the same. i'm going to try and find a way to force open the sealing between the trunklid and the car's body so that air can pas out of the seem at the top.
i hope this will cause air to stay attached to the rear window as is slopes down.
and thereby reduce the overall wake of the car