Aerodynamics question

[lbsf1]

Hi guys, wondering if any of you have any experience or knowledge of aerodynamics.

Basicly I'm part of a project at my school called greenpower where we build and race battery powered car in 4 hour endurance races, the racers are mostly just about going at a steady speed for that 4 hours, the speed we do is roughly 30mph.

Now comes the aerodynamics question, what is better for the bodywork of the car, a curved front or a pointy front. We currently run a curved front however I think a point (more deltawing like) would be an improvement. We are just looking for the best aero efficiency, no downforce or anything.

Also I was wondering whether openwheel or closed wheel would be better, since the wheels are thin I thought the effect of drag from them may be less, so the overall reduction in frontal section may in this case be better than the drag caused by wheels. Whats your view, open or closed, what would be better??

Here is what the car currently looks like, (About the first half of the vid is the making of the car (so you can see what the chassis looks like as a limitation for bodywork), then its pics from the race)

 

[dhandeh]

We had a similar project at work a couple of years ago, we decided the aero was secondary to getting the machine as light as possible. We went for a curved but somewhat pointed front end. As long as isn't Lada shaped you probably won't see much difference between the two.

 

[Danoff]

Closed wheel is superior for aerodynamics. But... at 30mph none of it matters much. You're just not moving through enough air to actually have substantial drag. As dhandes above said, focus on ditching weight and making it work reliably. Whatever you do do not add a spoiler. At 30mph it will give you nothing.

 

[Lizard]

At 30 mph aero has very little effect so the weight of any aero parts would be worse than the gain of the part.

 

[Slick Rick]

In the report Id mention that you decided not to focus as much on aerodynamics because of the low speed and instead talk about reducing weight and drag to make the most out of the motors and batteries....

 

[lbsf1]

Closed wheel is superior for aerodynamics. But... at 30mph none of it matters much. You're just not moving through enough air to actually have substantial drag. As dhandes above said, focus on ditching weight and making it work reliably. Whatever you do do not add a spoiler. At 30mph it will give you nothing.

Yeah, I'm not letting anyone get anywhere near putting a spoiler on the car, it doesn't have to do many turns anyway so that is far from a priority.

The car doesn't accelerate much faster than 30mph because of drag, the winner of the competition for the last couple of years wrote a guide saying aero was important, thats why I was wondering which way was most efficient.

I'm trying to lose as much weight of it as possible however it is difficult because the chassis is steel (aluminium one in construction). Hence would open wheel design decrease weight enough to outweigh its aero deficency.

 

[Lizard]

Yes I would think so. If your getting 30 mph due to drag do you have to drive it through water or something? Otherwise there is like 1 or 2 mph to gain from Adrien Newey designing the car.

 

[Famine]

Now comes the aerodynamics question, what is better for the bodywork of the car, a curved front or a pointy front. We currently run a curved front however I think a point (more deltawing like) would be an improvement. We are just looking for the best aero efficiency, no downforce or anything.

Wikipedia: Drag coefficients​

It's little accident that all solar cars and many high-efficiency concepts look like a teardrop with a very long tail...

Here's the most efficient car in the world:

It has a drag coefficient of 0.075 - a quarter that of a Toyota Prius.

 

[lbsf1]

Yes I would think so. If your getting 30 mph due to drag do you have to drive it through water or something? Otherwise there is like 1 or 2 mph to gain from Adrien Newey designing the car.

No not through water or anything, just around an airfield for 4 hours.

 

[dhandeh]

Yeah, I'm not letting anyone get anywhere near putting a spoiler on the car, it doesn't have to do many turns anyway so that is far from a priority.

The car doesn't accelerate much faster than 30mph because of drag, the winner of the competition for the last couple of years wrote a guide saying aero was important, thats why I was wondering which way was most efficient.

I'm trying to lose as much weight of it as possible however it is difficult because the chassis is steel (aluminium one in construction). Hence would open wheel design decrease weight enough to outweigh its aero deficency.

It's going to be a case of getting some testing done to see which works best. Get the tyres as thin as possible, put the pressure to maximum. The cockpit seems quite open, so maybe a section of bodywork could be fitted to cocoon the driver, a bit like in F1.

 

[Nessy]

I think you should definitely go with some shrouded wheels, similar to the 1's in the pic Famine has posted, or even teardrop shape shrouds maybe.

 

[corporatesteve]

As a pilot, aerodynamics will not play a huge roll at that speed when it comes to shapes. Its more going to be about material and what type of drag are affecting your surface. Also you have to look into whether adding aerodynamic properties will hinder performance due to weight and friction. Even with a shape that has a lower coef of drag could be the wrong setup. You have to research how that airflow gets directed, speed up, and disrupted with that shape your using. Skin friction drag will probably be your biggest enemy. Streamline the **** out of your ride, even to the helmets used. Look into laminar airflow.

 

[lbsf1]

I think you should definitely go with some shrouded wheels, similar to the 1's in the pic Famine has posted, or even teardrop shape shrouds maybe.

Yeah thats something that we were definately going to do, we tried it last year but they wouldn't fit properly.

We were also going to make a front windscreen to try to and push the air smoothly over our heads. However the rules don't allow anything over eye height.

Thanks for the image famine, that table is pretty useful and explains lots of eventualities for comparison.

 

[ExigeEvan]

The front is of less importance, the real gains will be in the rear of the vehicle, that's where the bulk of the turbulence will be and where the greatest gains will be made. I agree with many that drag is a minor issue at low speeds, but if any then frontal area is probably your biggest enemy rather than turbulent flow at those speeds.

But I'd focus on drive train efficiencies. Rolling resistance, drive train losses to friction, optimising engine speed.

Try hitting these guys up with a thread. They live for this stuff!
http://ecomodder.com/forum/ecomodding.html

[EDIT] I forgot to mention, smooth hubcaps with a slight convex. Ecomodders swear by them...

 

[Bopop4]

I'd shroud the rear wheels.
If you can still get clearance for the fronts then shroud them too.

 

[Danoff]

I still think drag is going to be a minor factor, but if you want to go nuts with it, see if you can come up with a way to manufacture body panels with dimples on them like a golf ball. Golf ball dimples help maintain a turbulent boundary layer around the surface of the ball. Mythbusters tested this on cars and found that it works there as well.

The next thing to do is come up with a few mockups of the design and take it to a wind tunnel to measure drag and figure out which design is best. Your local university aerospace engineering department has a wind tunnel that they'll let you use if you tell them it's for educational purposes.

 

[ExigeEvan]

The next thing to do is come up with a few mockups of the design and take it to a wind tunnel to measure drag and figure out which design is best. Your local university aerospace engineering department has a wind tunnel that they'll let you use if you tell them it's for educational purposes.

Get with the 21st Century gramps

 

[Danoff]

Get with the 21st Century gramps

Bad idea? You're thinking 3D modeling and simulated airflow? Might be a tad more expensive.

 

[homeforsummer]

I still think drag is going to be a minor factor, but if you want to go nuts with it, see if you can come up with a way to manufacture body panels with dimples on them like a golf ball. Golf ball dimples help maintain a turbulent boundary layer around the surface of the ball. Mythbusters tested this on cars and found that it works there as well.

I wouldn't entirely trust mythbusters on that one. I recall an F1 designer was once asked whether it would improve aerodynamics, and they said "No".

However, they did say it *might* be able to make a difference on smaller parts, like wing mirrors. But even then, the golf ball's dimples help turn an un-aerodynamic shape - a ball - into a more aerodynamic one. A smooth teardrop is still a much more aerodynamic shape, and as little turbulence as possible is required to maintain laminar airflow.

Worth also mentioning that in something as small as a car like lbsf1 is suggesting, frontal area is as important as the drag coefficient. There may not be much turbulence at 30mph, but any drag there is will be increased exponentially according to how big the vehicle is.

In other words, if you can find a way to cover up or reduce the effect of unaerodynamic objects - like the driver - then all's the better.

 

[Danoff]

I wouldn't entirely trust mythbusters on that one. I recall an F1 designer was once asked whether it would improve aerodynamics, and they said "No".

Well you don't want it on parts that actually perform an aerodynamic function - like creating downforce. You also don't really want them near surfaces that perform an aerodynamic function. So they'd be counter-productive on say... airplanes.

But for something that doesn't have any airfoils on it (unlike an F1 car), it could pay some dividends. Tough to manufacture though.

 

[ExigeEvan]

Bad idea? You're thinking 3D modeling and simulated airflow? Might be a tad more expensive.

Depends.

Wind tunnel time is generally quite expensive, combined with second and third visits to validate changes and it really adds up.

CFD wise you can start off with the simple, there's a variety of quite simple freeware 2D flow simulators which can help with general profiles, and there are freeware 3D flow simulators as well, but the UI are generally a *****. Not expensive though, as freeware, and computer demands are huge either.

Worth also mentioning that in something as small as a car like lbsf1 is suggesting, frontal area is as important as the drag coefficient. There may not be much turbulence at 30mph, but any drag there is will be increased exponentially according to how big the vehicle is.

In other words, if you can find a way to cover up or reduce the effect of unaerodynamic objects - like the driver - then all's the better.

Except the drag at 30 mph would most likely be negligible compared to the increased weight of aerodynamic parts required. Perhaps if they're quite adept at laying thin GRP bodies it may pay off, but it's no as simple as body work = better.

 

[ExigeEvan]

[EDIT] Sorry!

 

[Danoff]

Depends.

Wind tunnel time is generally quite expensive, combined with second and third visits to validate changes and it really adds up.

Wind tunnel time is likely free at the local university. We had a wind and water tunnel when I went through the aero engineering department, and it cost nothing to use.

 

[lbsf1]

Thanks guys, looking into the free cfd, and possibly seeing if surrey uni has a wind tunnel.

Gonna carry on with it tommorow, god getting up at 5.45 for a paperround is a killer.

 

[corporatesteve]

I wouldn't entirely trust mythbusters on that one. I recall an F1 designer was once asked whether it would improve aerodynamics, and they said "No".

However, they did say it *might* be able to make a difference on smaller parts, like wing mirrors. But even then, the golf ball's dimples help turn an un-aerodynamic shape - a ball - into a more aerodynamic one. A smooth teardrop is still a much more aerodynamic shape, and as little turbulence as possible is required to maintain laminar airflow.

Worth also mentioning that in something as small as a car like lbsf1 is suggesting, frontal area is as important as the drag coefficient. There may not be much turbulence at 30mph, but any drag there is will be increased exponentially according to how big the vehicle is.

In other words, if you can find a way to cover up or reduce the effect of unaerodynamic objects - like the driver - then all's the better.

F1 Designers dont know ****. They dont have the same research capabilities as aeronautical corporations do such as Boeing. Its been tested to aerodynamically be more efficient no matter what the size. Ive even met the lead aerodynamicist for Top Flight golf who is also the lead aerodynamicist for General Atomics producing the latest reaper and hes been the leading authority on this matter. The dimple design helps keep the airflow attached to the airfoil ( or surface ) longer than laminar flow and while airspeed increase( or in this case GS ) the drag stays the same ( where as it would normally have an increase in parasitic drag ). Obviously the dimple design is the basic concept for sporting equipment since there are other, more advanced theories that are being used in the field today. Formula 1 is more worried about disrupting the airflow and getting by loopholes. They could care less about real advances in aerodynamics.

 

[Keef]

As others have said, a teardrop shape will be your best bet. At 30 mph the benefit of a long, pointed nose will be almost non-existent, besides the weight penalty from the longer bodywork. Make sure the rear of the car either tapers to somewhat of a point, or is cut off at a sharp angle called a kammback. The kammback will help preserve a teardrop shape to the airflow while reducing the weight of the body structure. Also, the car won't need to be 20 feet long to complete the teardrop body shape.

The rear of the car Famine posted appears to end in a vertical teardrop which was made possible by the single rear wheel. Speaking of single wheels, you only need three to keep it upright. Three wheels make less drag and less weight than four.

I still think drag is going to be a minor factor, but if you want to go nuts with it, see if you can come up with a way to manufacture body panels with dimples on them like a golf ball. Golf ball dimples help maintain a turbulent boundary layer around the surface of the ball. Mythbusters tested this on cars and found that it works there as well.

The next thing to do is come up with a few mockups of the design and take it to a wind tunnel to measure drag and figure out which design is best. Your local university aerospace engineering department has a wind tunnel that they'll let you use if you tell them it's for educational purposes.

The dimples would only be useful at very high speed like what a golf ball flies at. They reduce friction with the air by exposing less surface area to it. It's the same principle you'll find on submarine skin and shark skin, reducing surface friction, but since air is so much less dense than water it will only work at high speed where the air can't fill low pressure areas quickly enough.

 

[Danoff]

The dimples would only be useful at very high speed like what a golf ball flies at. They reduce friction with the air by exposing less surface area to it.

Not exactly. They actually increase the amount of surface area exposed to the air. What they do is convert the laminar boundary layer to turbulent

With a turbulent boundary layer, the air flow moves around the boundary layer surrounding the ball rather than around the ball itself reducing the turbulent wake left by the ball

It's the same principle you'll find on submarine skin and shark skin, reducing surface friction, but since air is so much less dense than water it will only work at high speed where the air can't fill low pressure areas quickly enough.

I'm not sure what the threshold for this technique working is. But I suspect that if you launched a golf ball at 30mph, one with the dimples would go farther than one without.

Well you don't want it on parts that actually perform an aerodynamic function - like creating downforce. You also don't really want them near surfaces that perform an aerodynamic function. So they'd be counter-productive on say... airplanes.

Upon further thought, I'm not entirely sure this is the case. Seems like an airfoil should work just fine with a turbulent boundary layer.

 

[Omnis]

Is there some kind of optimization for the size of the dimples relative to the curve? It seems like that's something that should be on all moving vehicles, but the math's probably too hard to calculate every model year.

 

[Keef]

Not exactly. They actually increase the amount of surface area exposed to the air.

Dimples decrease the surface area exposed to the airflow. This reduces friction with the air, the friction being what slows the air and increases the boundary layer thickness which causes flow separation.

On the front surface of the ball the reduced surface area takes immediate effect, with the air flowing over air trapped inside the dimples as opposed to flowing over what would be more high-friction surface area. This reduction in surface area causes the reduction in friction buildup which extends the separation point. The benefit of the smaller wake is less parasitic drag.

EDIT:

Somebody call an aerodynamicist.

This is all I know:

But I believe vortex generators work by a different mechanism.

 

[Azuremen]

Wikipedia: Drag coefficients​

I took the liberty of amending the figures to more accurate values.