Who uses vortex generators?

[ryannoe]

Originally Posted by caferacer51

I was under the impression that the VG on the rear roofline of the Lancer was to aid rear down force? Surely that goes against this conversation of VG use for economy?

And in the game of fuel economy, is 2% not a substancial gain?

They were used to improve downforce... What happened in the EVO's circumstance is flow separation decreased the authority the spoiler had. No flow over the spoiler.

The vortex generators were brought in to push the separation point lower down the back glass which helped pull some freestream air towards the spoiler.

I'd take 2% over nothing

-Ryan

[AeroModder]

http://ecomodder.com/forum/showthrea...tml#post294139

[RedDevil]

Originally Posted by ryannoe

RedDevil - unless you have introduced a new surface, I have my doubts that it is laminar at your rear bumper. Laminar flow is more likely to separate from the vehicle than turbulent flow (which I suspect is what you have). I'd like to hear what happens. Don't forget to post it for us!

-Ryan

To be precise: my point is that the airflow on the side bumper is usually turbulent unless that bumper is hit by strong sidewind.

I don't know for sure as I have no way to see what actually happens.
I just deduced what happens from the bits of information I have:

Observation
- The 2Gen Insight suffers from side wind effects; under some conditions I can feel a constant but erratical series of jerks, esp. when moving slightly with the wind instead of against it.

- Fuel consumption is much higher on side wind impact than when driving straight against the wind. At 6 Bft I noticed the FC went up 30% compared to driving straight against it driving on a gently sweeping highway in a completely flat and almost treeless plane, steady pace etc.

Theory
- Airflow on a plate that leans into the wind tends to be laminar.

- Airflow moving along a plate is unstable and tends to get turbulent.

- Rounded edges can make an airflow passing along it follow the curve. This is the Coanda effect. A laminar flow may follow the curve completely.

- The Coanda effect is inherently instable and tends to enlarge any instabilities in the airflow until the turbulence allows it to break loose.
The radius of the bend is strongly related to the instability; if it is too sharp the flow cannot follow it and breaks loose.

- Turbulent air won't follow an inward curve but rather move on in the same direction.

- Vortices are more or less stable by themselves and folllow a curved edge, especially the part of the vortex that is already moving in that direction.
That may explain why vortices help in the Mitsubishi case; it allows part of the air to curve an edge too sharp for laminar or turbulent air to follow.

Deduction
When hit by a strong sidewind the airflow curves around the edge of the rear bumper, creating a low pressure area at the side and rear of the bumper, pulling it back.
Straight against the wind the air is turbulent and ignores the corner when it moves by.
In between the air will sometimes curve round and sometimes go straight. Both conditions are unstable and likely to induce the other. This produces the shaking sensation.

Remedy
Breaking or preventing the Coanda effect on the corner should enhance stability and yield better MPG.
One way is to sharpen the trailing edge (Prius II, Chevy Volt etc.)
Another is inducing turbulence by adding a ridge or crease or sticking on zigzag tape just before the corner. The latter is my plan.

Proof
Nobody that I know of has done this, however Cobb has removed his rear bumper and claims that improves side wind stability and MPG.
The uncovered bodywork falls back slightly with a sharp cliff and has sharp edges.
This is no way as much proof as I would like, but it does fit.

Testing
Both my car and me are quite new to this game. There are some more mods that I want to do and some testing gear to get and use. Especially the rear wheel well covers, as I suspect that will decrease turbulence and so increase the effect of the edge curve cure.

Once that is sorted out and the conditions are right (I need strong and stable southwestern wind, like 6 or 7 Bft) then sure I will get out the word.

Meanwhile, if there's anyone out there with a wind tunnel and an Insight, I'd be over the moon if you could park it slightly askew in that tunnel and test it with and without zigzag tape, ridges and the like. Well, who knows?

[ryannoe]

Originally Posted by RedDevil

Theory
- Airflow on a plate that leans into the wind tends to be laminar.

- Airflow moving along a plate is unstable and tends to get turbulent.

- Rounded edges can make an airflow passing along it follow the curve. This is the Coanda effect. A laminar flow may follow the curve completely.

- The Coanda effect is inherently instable and tends to enlarge any instabilities in the airflow until the turbulence allows it to break loose.
The radius of the bend is strongly related to the instability; if it is too sharp the flow cannot follow it and breaks loose.

- Turbulent air won't follow an inward curve but rather move on in the same direction.

- Vortices are more or less stable by themselves and folllow a curved edge, especially the part of the vortex that is already moving in that direction.
That may explain why vortices help in the Mitsubishi case; it allows part of the air to curve an edge too sharp for laminar or turbulent air to follow.

I don't doubt your observations and I'm mulling over why I think you're experiencing the negative MPG.

However, I think you have turbulent and laminar flow flip-flopped. Laminar flow will separate early while the turbulent air tends to 'stick' a little longer. Its hard to think about and conceptualize... [been sitting here backspacing for about 5 minutes trying to think of a good way to explain it]

The usual example is the golf ball but I don't like that. With turbulent air, the momentum is broken and skin drag increases. The particles have vectors in all directions which means SOME of them have vectors in the direction we WANT it to go. In laminar flow, the air is saying "No sir, me and my buddies are going this way."... However since they are all going in that direction, there is less air bouncing off of the surface causing less skin friction. Finesse is needed to keep laminar flow as long as needed and then trip the flow to turbulent so it will stick to the converging tail.

Vortices are such an interesting subject matter. The low pressure region in the center wants to suck the air inward, while the centrifugal force wants to pull it apart... while the whole time, the shear forces within caused by speed differentials cause eddies and reverse vortices (can you tell I geek over vortices? ). Not sure why, and it's a high level subject, but the center of the vortex DOES have an affinity to a surface. I doubt this is evident in the dirty tails of our vehicles...

Ok, sorry so long. Have a great night!

-Ryan

[RedDevil]

Originally Posted by ryannoe

However, I think you have turbulent and laminar flow flip-flopped. Laminar flow will separate early while the turbulent air tends to 'stick' a little longer. ...

You know, I've been wondering about the Coanda effect and its limits.
I may as well have misjudged the role of laminar flow and turbulence. I can't be certain. But I find the Coanda effect hard to explain with anything other than laminar flow.

The only thing that can make a stream of air or water alter its direction is a difference in pressure (and vice versa).
I can visualize that happening in laminar flow. I cannot visualise that in turbulent flow. Parts of a turbulent flow will follow the bend but as it requires less energy to just move on, that's what most of the air will do.

But I do admit that I worry about the effect of zigzag tape on the corner to break the Coanda effect. A sharp edge will break the effect but that does not depend on the properties of the airflow (being laminar or turbulent). The zigzag tape however only works if my argument is true and yours is not. Which makes my alarm bells ring. Or, scientifically; Hey, this might be interesting; what is really happening here?

[Sven7]

I think there's a vortex generating in this thread...

[ryannoe]

RedDevil - I'm treading lightly around the Coanda effect. I believe the "Coanda effect" is really just a case of Bernoulli's principle without the duct. You can't have it without there being a jet present. Please don't believe 99% of what you read on the web about it, because it looks like it's coming from those "scientists" that "don't need a formal education". You know, the ones that think you can put a wind turbine on a car and drive for free. At least those were the sites Google pulled up for me.

Take a piece of paper and hold one end of it. Let the other end dangle. Blow strongly across the top surface of the paper and it will lift. This is because the static pressure in the jet is less than the static pressure of the ambient air. The air between the paper and your blown jet is sucked up into the jet and thus lifting the paper. This principle explains why the wake pressure behind flow separation is governed by the low pressure of the separation point. The air is being sucked back up to the low pressure point at the separation (flow reversal...).

BACK TO YOUR CAR AND CROSSWIND...
I'd suspect the influence to your vehicle's aerodynamic AoA is causing induced drag but the big culprit is probably the incidental lift. Although the lift is perpendicular to your path (so it shouldn't matter right?), you're having to compensate for that perpendicular force by correcting your steering. I bet your MPG losses are due to your now misaligned wheels rubbing the road to make up for your aerodynamic force vector.

Bed time! Have a great night!

-Ryan

[ChazInMT]

I think people tend to lose sight of the big picture when talking about VG's and Ziggy Tape. The air at the skin of the car is only a tiny fraction of the air that is being affected by the passing of the vehicle, certainly the air at 3 to 4 feet out from the sides of a vehicle is fully in play, and for many tens if not hundreds of feet behind. It is only by gross management of this air flow that we can expect significant gains, meaning changing the shape of our cars. Screwing around with the already "Turbulent Boundary Layer" is a piffle compared to the huge amounts of air that are important to what the overall Cd is on the car.

This is my primary reason for saying VG's & ZT's on cars are a waste of time when trying to achieve any real gains. Fun to think about I suppose, but in the end you need to visualize the thousands of cubic feet of air that are moving up, down, to and fro, and ask yourself if something sticking up a few inches tops is really gonna do anything.

Clearly the VG's & stuff are great for airplanes, I understand that. But again it is because you're dealing with airflows on wings which is laminar by nature, car aero is seldom laminar.

[ConnClark]

Originally Posted by ChazInMT

Clearly the VG's & stuff are great for airplanes, I understand that. But again it is because you're dealing with airflows on wings which is laminar by nature, car aero is seldom laminar.

My CFD simulations show laminar flow along the majority of the body of a car.

Furthermore, the boundary layer has huge effects on flow separation. Reduction of flow separation in certain areas of a car can reduce the C pillar vortices strength. This reduction can greatly lessen the size of the wake of a car and thus reduces the induced drag.

[ryannoe]

Originally Posted by ConnClark

My CFD simulations show laminar flow along the majority of the body of a car.

Furthermore, the boundary layer has huge effects on flow separation. Reduction of flow separation in certain areas of a car can reduce the C pillar vortices strength. This reduction can greatly lessen the size of the wake of a car and thus reduces the induced drag.

Exactly.

What CFD software are you using?

-Ryan