effects of a roof wing?

[Grip Ronin]

Originally Posted by plucas

Flow stays attached to rear glass pretty well without anything. No vortex generators needed. I could make some assumptions about a roof spoiler, but it would just be an assumption.

The picture below shows how the fluid stays attached

thats a good picture, what would be your assumption of the roof spoiler?

[Gunman]

CFD explains more than my speculation would, but that said....roof spoiler *could* work as a trip, to get the air reattached, although CFD seems to show it is already attached.

and OP, that's not a roof "wing", it's a spoiler. Wing's have air flowing under them.

[Calidrifter]

I believe that it would allow less air to get into the trunk via the gap between the glass and the trunk. TRD addressed this issue with an added piece at the base of the glass.

[soundman98]

Originally Posted by plucas

Flow stays attached to rear glass pretty well without anything. No vortex generators needed. I could make some assumptions about a roof spoiler, but it would just be an assumption.

The picture below shows how the fluid stays attached

would you like to borrow a roof spoiler for some further testing? or is what you do all computer generated models (making the borrowing of a physical wing kinda useless)?

[Grip Ronin]

that would be a very helpful test^^. even though the roof spoiler has a small lip, i feel it would toss the air up and over the edge of the trunk so i may just rip it off even though i like it

[kiichiro]

Earth is not flat?

[soundman98]

it is if you put your face against it and sight into the distance

[plucas]

Originally Posted by Grip Ronin

thats a good picture, what would be your assumption of the roof spoiler?

I would assume it would decrease airflow to the rear lip. I do not know if it would have a net negative or positive lift numbers though. Really just depends.

Originally Posted by soundman98

would you like to borrow a roof spoiler for some further testing? or is what you do all computer generated models (making the borrowing of a physical wing kinda useless)?

Most of my testing is computer based (CFD). I do however try and verify with real world testing when possible.

Roof spoiler would just allow me to model the spoiler accurately. Then I could test how it effects the airflow.

[chrisl]

Originally Posted by plucas

Flow stays attached to rear glass pretty well without anything. No vortex generators needed. I could make some assumptions about a roof spoiler, but it would just be an assumption.

The picture below shows how the fluid stays attached

Just to clarify what's going on in that image, the colors in the fluid domain are all based on velocity, not pressure, right? And the colors on the car surface itself are pressure? I'm just trying to make sure I'm interpreting this picture correctly - when I first saw it, I was scratching my head trying to figure out why on earth there would be a high pressure region on top of the roof, but that area makes a lot more sense as a low pressure, high-velocity region.

As for what the roof spoiler does? My suspicion (to which I will apply all of the same disclaimers as everyone else - I'm just guessing here, and I could be way off base) is that it will cause the flow to detach at the top of the rear window, and thus you'll get a region of separated flow over the back of the car. This would probably decrease the car's lift at high speeds, especially for the back end, but it would also make any rear spoiler useless unless it protruded out pretty far (into the clean airflow above the recirculation zone). It would also probably increase drag significantly. Based on the CFD image, it looks like the airflow on the unmodified car is slowing down in that region (since it looks like it hits its peak speed on top of the roof), so there is probably an adverse pressure gradient there that would prevent the flow from reattaching after a forced separation.

That having been said, it is also possible that the separation caused by the roof spoiler isn't dramatic enough to cause a full flow separation from the rear window, and instead you would just get a small recirculation bubble behind the spoiler and then the flow could reattach itself to the rear window. This is definitely a possibility, despite the adverse pressure gradient that is probably present, since the spoiler isn't at a very dramatic angle, and the step off the back of the spoiler onto the glass is pretty small. In that case, it really wouldn't do much for anything other than appearance.

Interestingly, if my first guess is correct, this is doing pretty much the exact opposite of what vortex generators do (such as the ones commonly seen on Evos/STIs on the back of the roof). The vortex generators are used to try to energize the boundary layer and get it to attach better to the relatively sharply sloped rear window of a sedan (and preventing flow separation), thus reducing the drag and improving the airflow over the trunk (and any trunk mounted aero aids, such as a spoiler). This device would increase the boundary layer thickness just after it and ensure that the airflow next to the surface is relatively stagnant. This could cause an otherwise nice, smooth attached flow from the roof onto the gently sloped rear window of a coupe to separate, increasing the drag and harming the airflow over any trunk mounted aero.


Of course, this is all speculation without either wind tunnel/flow visualization results, or a nice, high-fidelity computer simulation (preferably one which is good at resolving turbulence, boundary layers, and flow separation, which can be surprisingly difficult sometimes).

[plucas]

Originally Posted by chrisl

Just to clarify what's going on in that image, the colors in the fluid domain are all based on velocity, not pressure, right? And the colors on the car surface itself are pressure? I'm just trying to make sure I'm interpreting this picture correctly - when I first saw it, I was scratching my head trying to figure out why on earth there would be a high pressure region on top of the roof, but that area makes a lot more sense as a low pressure, high-velocity region.

As for what the roof spoiler does? My suspicion (to which I will apply all of the same disclaimers as everyone else - I'm just guessing here, and I could be way off base) is that it will cause the flow to detach at the top of the rear window, and thus you'll get a region of separated flow over the back of the car. This would probably decrease the car's lift at high speeds, especially for the back end, but it would also make any rear spoiler useless unless it protruded out pretty far (into the clean airflow above the recirculation zone). It would also probably increase drag significantly. Based on the CFD image, it looks like the airflow on the unmodified car is slowing down in that region (since it looks like it hits its peak speed on top of the roof), so there is probably an adverse pressure gradient there that would prevent the flow from reattaching after a forced separation.

That having been said, it is also possible that the separation caused by the roof spoiler isn't dramatic enough to cause a full flow separation from the rear window, and instead you would just get a small recirculation bubble behind the spoiler and then the flow could reattach itself to the rear window. This is definitely a possibility, despite the adverse pressure gradient that is probably present, since the spoiler isn't at a very dramatic angle, and the step off the back of the spoiler onto the glass is pretty small. In that case, it really wouldn't do much for anything other than appearance.

Interestingly, if my first guess is correct, this is doing pretty much the exact opposite of what vortex generators do (such as the ones commonly seen on Evos/STIs on the back of the roof). The vortex generators are used to try to energize the boundary layer and get it to attach better to the relatively sharply sloped rear window of a sedan (and preventing flow separation), thus reducing the drag and improving the airflow over the trunk (and any trunk mounted aero aids, such as a spoiler). This device would increase the boundary layer thickness just after it and ensure that the airflow next to the surface is relatively stagnant. This could cause an otherwise nice, smooth attached flow from the roof onto the gently sloped rear window of a coupe to separate, increasing the drag and harming the airflow over any trunk mounted aero.


Of course, this is all speculation without either wind tunnel/flow visualization results, or a nice, high-fidelity computer simulation (preferably one which is good at resolving turbulence, boundary layers, and flow separation, which can be surprisingly difficult sometimes).

Correct. The cut plot is velocity and the body surface is pressure.

As for your second and third paragraph, I agree.

[Grip Ronin]

hm im thinking the same, may be time to sell the roof wing if i want the possible downforce of the trunk lip

[kidgogeta]

Originally Posted by Grip Ronin

hm im thinking the same,may be time to sell the roof wing if i want the downforce of the trunk lip

lol ill take it off your hands

[7thgear]

Originally Posted by plucas

Flow stays attached to rear glass pretty well without anything. No vortex generators needed. I could make some assumptions about a roof spoiler, but it would just be an assumption.

The picture below shows how the fluid stays attached

is your graph a result of a windtunnel or just graph plots?




an acquaintance of mine did his own testing on a 2nd gen rx-7 with strings taped to the rear glass. without any form of diffuser they would be turbulent, with a diffuser they become uniform.


how close/smooth does the air flow over "sport coupes", pretty well I suppose, but from what I understand, the use of small diffusers still offers an improvement.

[plucas]

Originally Posted by 7thgear

is your graph a result of a windtunnel or just graph plots?




an acquaintance of mine did his own testing on a 2nd gen rx-7 with strings taped to the rear glass. without any form of diffuser they would be turbulent, with a diffuser they become uniform.


how close/smooth does the air flow over "sport coupes", pretty well I suppose, but from what I understand, the use of small diffusers still offers an improvement.

I guess I do not understand. This was done with Computational Fluid Dynamics (CFD) which solves numerical equations fore fluid flow.

Diffusers can help flow above the car, flow before the car, and behind the car.