A question on vehicle aerodynamics

[grey]

ok...i have a small question on aerodynamics for cars:

in fluid mechanics, we studied the flow of air past an airfoil, which is designed to reduce drag (and obviously lift has a lot to do with it, but that is irrelevant for me now, so i am ignoring it). My question is that why aren't cars designed that way? that is, 'fatter' at the front and 'sleek' at the back? in fact, i seem to think that they are designed in the opposite way...why is that?

 

[mgb_phys]

Because you have to fit people and an engine in them.
Ultra economy racing (where people go for miles on a teaspoon of fuel) are shaped like a teardrop , but for most cars it is a question of rounding off a square box a little and avoiding a trailing vortex.

 

[grey]

hi, thanks for the reply

actually the 'ultra economy racing' is what we intend to do and that made me ask
a further question...is the tear-drop shape effective as compared to the opposite-tear-drop shape at low speeds? by low, i mean something like 30 kph (around 8 m/s)?

 

[Cyrus]

ok...i have a small question on aerodynamics for cars:

in fluid mechanics, we studied the flow of air past an airfoil, which is designed to reduce drag (and obviously lift has a lot to do with it, but that is irrelevant for me now, so i am ignoring it).

Who taught\told you this? Airfoils are designed to optimize lift to drag ratios.

My question is that why aren't cars designed that way? that is, 'fatter' at the front and 'sleek' at the back? in fact, i seem to think that they are designed in the opposite way...why is that?

They're not. Car's are designed with streamlining in mind. Read [1]:

[1] http://en.wikipedia.org/wiki/Kammback

 

[minger]

The tear-drop shape is effective because it is optimum at keeping the boundary layer attached. As flow slows down and pressure increases, we refer to it as an adverse pressure gradient. This can lead to boundary layer separation, which hurts drag.

The idea is that if I'm going to sit in a car, then it has to be "X" high...somewhere. So, put it towards the front, keep the favorable pressure gradient are small, and then SLOWLY reduce the thickness as to ensure the BL stays attached.

If you check out:
http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19940008699_1994008699.pdf
you can see the flow separating just past the turning point. That's because the adverse pressure gradient is too high. Reduce the change in thickness and reduce that gradient. That's the idea.

 

[xxChrisxx]

The reason why cars arent like that is because they look stupid, and the person driving it looks stupid. Customer tastes and marketing are the primary reasons for why cars arent streamlined bodies.

Opposite teardrop shaped arent effective at all, for cars the vast majority of the Cd is from pressure drag caused by separated flow. (This reminds of the Austin allegro that was acutally more aerodynamically efficient in reverse). Having a wide bit at the back, casues the gradient to be too large and you get separated flow.

If you want to go for ultra economy, a long car that is teardorp shaped where the drvier lies down will be the best.

 

[Cyrus]

The reason why cars arent like that is because they look stupid, and the person driving it looks stupid. Customer tastes and marketing are the primary reasons for why cars arent streamlined bodies.

Cars are like helicopters. The unaerodynamic shapes they assume are a result of needing to have utility. If you get a chance to drive a kammback car, you quickly realize the roofline on the back seats is very low, and the shallow slope of the roofline leads to large blind spots and horrible rear view visibility. Not only is inside cabin space reduced, the car becomes a safety issue when it's tear drop shaped.

 

[xxChrisxx]

Be that as it may. Bottom line is for whatever reasons you can come up with (and yours were good ones). Cars aren't teardrop shaped because people won't buy them.

 

[Cyrus]

Be that as it may. Bottom line is for whatever reasons you can come up with (and yours were good ones). Cars aren't teardrop shaped because people won't buy them.

It's not a matter of 'people wouldn't buy one', it's that the vehicle would not have any utility. If it got 100MPG, people would absolutely buy a teardrop shaped car. It would also go faster. I see no justification as to why you think people would not buy one. People buy american cars, don't they? QED.

 

[grey]

ok...thanks guys for the views...here is a design i made on google sketch-up
it s a bit crude, and no dimensions are given, but if you can critique i would be over-joyed
also, tips, ideas for further improvement are most welcome

and yeah, the prospected speed (max) is around 30 km/h (8.3 m/s)

 

[Ranger Mike]

that was a good one..( by American cars..).I had a good laugh on that one...the original question was " why aren't cars designed that way? that is, 'fatter' at the front and 'sleek' at the back?
Many versions of this have been prototyped and Automobile Marketing Departments have test marketed these designs in the public consumer marketplace for years..we do see the evolution trending toward more effective Areo designs..but the overwhelming marketing driver is ..Big Government, with EPA and CAFE Standards.. Big Brother dictates the MPG minimum and Big Auto moves mountains to get there..when this stuff hit the fan in the mid 1970's, the big Cube V8 engines got dumped..no more 426, 426, 440, 454, 455, 460 cubic inch displacement engines...then the CAFE regulations kicked in and the two tin big cars got on the Jenny Craig diet and weight was reduced.( so was crash safety but this was not hyped because it would make Big Gov. look bad)
thins settled out for a while until the green stuff became trendy and everyone was herded like sheep by the 'man made global warming" sham. Now General Motors ( owned d by the Government ) is making pregnant roller skates that will only be purchased by Federal, State and Local government offices and greenie weenie tree huggers who want to feel good abut them selves. The general public will not want these clown cars and the top down dictated product like will go the way of the Edsel,Gremlin, and Yugo.
So the bottom line is...given the confines of government regulations ...CAFE, EPA, DOT and others..how do we economically manufacture an automobile that will comply to all these regulations AND appeal to the purchasing public more than the off shore Competitor.
It is one big juggling act.

 

[xxChrisxx]

It's not a matter of 'people wouldn't buy one', it's that the vehicle would not have any utility. If it got 100MPG, people would absolutely buy a teardrop shaped car. It would also go faster. I see no justification as to why you think people would not buy one. People buy american cars, don't they? QED.

Because I along with many other drivers would rather look awesome in a flashy and impractical but good looking car.

So until someone designed a teardrop car that doesn't look...well dreadful frankly. Noone will want one. Even then, you have the practicality issues you mentioned.

As much as people may say they want a car that gets uber fuel economy, it could do a million miles to the gallon but if its shaped like a portable toilet no one will buy it. Nobody on this planet wants to be A: laughed at. B: known as 'the sensible one', its just not cool.

Also no one really cares what you think, no one really cares what I think. Its not on topic and will and up leading to us being stubborn and bickering. So as to not derail the thread shall we leave it there?

 

[xxChrisxx]

ok...thanks guys for the views...here is a design i made on google sketch-up
it s a bit crude, and no dimensions are given, but if you can critique i would be over-joyed
also, tips, ideas for further improvement are most welcome

and yeah, the prospected speed (max) is around 30 km/h (8.3 m/s)

Hmm... dimensions are really needed. It looks ok, but its a bit wide, you want the frontal area to be as small as possible.

You may also want to smooth it out a bit, edges generally = bad.

It's difficlt to make a proper judgement without running it through CFD. Also what about ride height?

 

[grey]

edges...point noted
wide...actually, its a very crude diagram...but max width is going to be around 1.2 m around the wheels and less than 1 m above
height...around 1.2 m max

and i read on some website, http://www.cardesignonline.com/design/aerodynamics/aerodynamics2.php
it says that 'drag due to rear slope angle would be at its peak at 30 degree, and min at 10 degrees'
is this correct?

there's no way i can do 10 degrees...max possible is around 25, which would, if the above statement is correct, increase drag

 

[xxChrisxx]

It's all to do with separation of air and form drag (so drag that comes from the shape, such as skin friction).

It's a trade off between, having separated flow (pressure drag) and skin friction (form/parasitic drag). A very short car with a high slope will have low form drag because there isn't much material, but will have high pressure drag due to separated flow. A very long car will have more form drag but little to no pressure drag.

This is the reason why very long lower speed trains can have blunt noses, the pressure build up is negligable compared to skin fricion. But a short high speed train will have a bullet shaped nose.What you want is the slope that does not let any flow separate, but is not longer than necessary.

If the best you can get is near to 30 degrees for a full tear drop, you may consider using a lower angle but cutting off the rear, like the Kamm tail car Cyrus was talking about.

The other option is to simply make the car longer or lower, obviosly if you are building to a spec then this may not be possible.

I can't really comment on specifics of this car because there simply isn't enough data to make a reasoned guess.

 

[Cyrus]

ok...thanks guys for the views...here is a design i made on google sketch-up
it s a bit crude, and no dimensions are given, but if you can critique i would be over-joyed
also, tips, ideas for further improvement are most welcome

and yeah, the prospected speed (max) is around 30 km/h (8.3 m/s)

All you drew was a shape... I see no tires, no engine, no drive train, no cabin, no trunk. And that's the point. Until you do some serious thinking and analysis what you provided isn't nothing more than a drawing. You will find that your cars shape will have to change drastically due to the subsystems, stresses, and kinematics needed to make the car work. Right now you're toying around. Come up with something serious and you're going to learn a lot about automobiles in the process. Go on websites and find out the specs of actual components and fit them into your design. It will give you an appreciation for engineering, which is drastically different from what you're doing - design.

 

[Ranger Mike]

Grey, you keep doing what you are doing!..I am positve you will do just fine...never let negative feedback kill off the inventive side of your idea making...keep posting and never let doom and gloomers get you down...now get back on the drawing board and send us all another good post/ CAD pic! We old guys need to be reminded of the refreshing, new , energized way to do things ...some times...go get um, Tiger!

 

[grey]

If the best you can get is near to 30 degrees for a full tear drop, you may consider using a lower angle but cutting off the rear, like the Kamm tail car Cyrus was talking about.

thanks again

ok...i think we ll have to use this kamm back concept...but does it make sense to put things that make flow turbulent at the place the kamm back ends? something like the dimples on golf ball? i saw in some video on youtube that some new car they designed (i don't remember who, sorry) it had small fin-like objects to do the very same thing?

All you drew was a shape... I see no tires, no engine, no drive train, no cabin, no trunk. And that's the point. Until you do some serious thinking and analysis what you provided isn't nothing more than a drawing. You will find that your cars shape will have to change drastically due to the subsystems, stresses, and kinematics needed to make the car work. Right now you're toying around. Come up with something serious and you're going to learn a lot about automobiles in the process. Go on websites and find out the specs of actual components and fit them into your design. It will give you an appreciation for engineering, which is drastically different from what you're doing - design.

you have a perfectly valid point, but at the moment, i am doing an internship...i have no one else working on this project; at the moment, i am on ground zero...so i am really sorry i am doing this like i am and putting you guys in trouble

what i had in mind was to have an idea of the aerodynamics, and hence the shape...then design the chassis according to the shape...then calculate the weight, etc needed and according to that i ll need the power source, which can be a simple engine or a wholee hybrid system depending on what i have and what i need

i ll try to do what you have told me nevertheless, thanks

 

[Ranger Mike]

before you get into vortex generators ...check out drag coefficents and the like..the whizzy bits like vortex generators, dimples etc .. come after the basics...i recommend
get up on the internet and research the following

Bernoulli law of constant pressure
flow characteristics with in boundary layer
aero dynamic drag formula
parasite drag
Cd numbers
air velocity and pressure curves for airfoils
wing tip vortices
ground effects

will give you pretty good start on figuring out air / aero etc..
As Waddel Wilson ( famous NASCAR crew cheif ) once said.." anytime you are moving, yer pushing AIR"

 

[xxChrisxx]

All you drew was a shape... I see no tires, no engine, no drive train, no cabin, no trunk. And that's the point. Until you do some serious thinking and analysis what you provided isn't nothing more than a drawing. You will find that your cars shape will have to change drastically due to the subsystems, stresses, and kinematics needed to make the car work. Right now you're toying around. Come up with something serious and you're going to learn a lot about automobiles in the process. Go on websites and find out the specs of actual components and fit them into your design. It will give you an appreciation for engineering, which is drastically different from what you're doing - design.

Ok, now this is on topic so I am free to fight about it.

Before you start being impertanent, read the original posts. This isn't a car to tool around town in, this is a car for an endurance racing project. It's perfectly acceptable to start by dsigning the outer cladding. Driver comfort and practicalites are secondary. It doesn't need a trunk and the rest of the stuff can be crammed in where ever. Sizes can even be increased if the rest of the stuff can't be packaged.

Also you've got to start somewhere with design, with an endurance racer, aero is as good as any. I didnt realize people needed a complete engineering project to post here.

The OP's ideas, questions and design starting point are perfectly valid. So unless you have something decent to contribute please stop filling the thread with drivel just for the sake of argument.

 

[xxChrisxx]

thanks again

ok...i think we ll have to use this kamm back concept...but does it make sense to put things that make flow turbulent at the place the kamm back ends? something like the dimples on golf ball? i saw in some video on youtube that some new car they designed (i don't remember who, sorry) it had small fin-like objects to do the very same thing?

Well obviously a full teardrop is ideal, but the advantage of a Kamm back is that it gives marginally more drag for a more practical design. You will get pressure drag as the flow will separate but you've got to decide if a Kamm tail of 10 degrees will be better or worse than a full teardrop at 25 degrees in terms of drag.

I'm not telling you what to do, or what will be best. Those design decisions will have to be yours in the end. I'm just giving a few options of how to achieve the goal.

 

[Cyrus]

thanks again

ok...i think we ll have to use this kamm back concept...but does it make sense to put things that make flow turbulent at the place the kamm back ends? something like the dimples on golf ball? i saw in some video on youtube that some new car they designed (i don't remember who, sorry) it had small fin-like objects to do the very same thing?


you have a perfectly valid point, but at the moment, i am doing an internship...i have no one else working on this project; at the moment, i am on ground zero...so i am really sorry i am doing this like i am and putting you guys in trouble

what i had in mind was to have an idea of the aerodynamics, and hence the shape...then design the chassis according to the shape...then calculate the weight, etc needed and according to that i ll need the power source, which can be a simple engine or a wholee hybrid system depending on what i have and what i need

i ll try to do what you have told me nevertheless, thanks

Hi Grey. As an engineer that's been on a design team before, I need to point some things out here that I don't see you doing (which is why I'm giving you grief). If you are serious about any form of design on a car, you really need to take a scientific approach. In any design process, you need to do some background literary research to see what's been done, and where there is room for potential innovation. Simply drawing a car and then saying such things as "ok...i think we ll have to use this kamm back concept...but does it make sense to put things that make flow turbulent at the place the kamm back ends" shows me you are making it up as you read posts. I would go to the library and find some SAE articles on automobile design. You don't need anything technical, but you do need to have some basic, realistic understandings of the systems needed to make the car work.

Also, any good design is made with to conform to constraints. As of yet, I see you have no such constraints. Pick an engine size, a maximum car length, number of passengers. You're drawing doesn't have anything as basic as a linear dimension in it.

Once you do some background research, you can make a serious effort at a drawing. I'm sorry to be blunt, but what you're doing right now is not engineering. I would hate to see you guys actually start building something on a whim, because that's how it comes off the more I read your posts. I don't see any engineering rationale.

 

[grey]

thanks for the kind words, and advice...

Ranger Mike...actually we have studied most of what you told me to look, but i looked up those things anyway...basically decided to go back to the basics...now a question that comes to my mind is that, why, as most resurces on the boundary layer say, do we need to keep it attached to the vehicle?
In the boundary layer, the viscous forces are dominant...so viscous drag increases?
i think that for cars, the pressure drag would have a dominant effect (at those speeds) so we would rather have viscous than pressure if and only if the overall drag is reduced...am i right here?

 

[Ranger Mike]

excellent question - the short answer is that we want to keep the boundary layer attached so we can manage it ( control it like we control sprung weight). We want the layer detaching past the end of the car. If portions of the air stream detach before this, we have created more drag than if is detaches after passing completely over the car. Same reason two race cars run faster while drafting ( slip stream).

 

[xxChrisxx]

You want the boundary layer attached for precisely the reason you said, pressure drag. Pressure drag is far more killing to overall Cd than skin friction is.

After the flow separates it creates a pocket of low pressure that wants to drag the car back, this combined with the high pressure pocket in front of the car is what casues the drag.

You may also think that nice smooth laminar boundary will be better, this isn't acutally true. A turbulent boundary layer is more energetic and will stay attached to the outer skin for longer.

It's important for you to know why the boundary layer seperates. Read up on adverse pressure gradient.

 

[grey]

excellent question - the short answer is that we want to keep the boundary layer attached so we can manage it ( control it like we control sprung weight). We want the layer detaching past the end of the car. If portions of the air stream detach before this, we have created more drag than if is detaches after passing completely over the car. Same reason two race cars run faster while drafting ( slip stream).

Noted!

It's important for you to know why the boundary layer seperates. Read up on adverse pressure gradient.

will do

Ok, while i have been studying a bit on one side, i tried to make the thing in AutoCAD...posting the gif here (dont know how to post dwg)...hope both of you guys can review it there
its still not there (things like the horizontal edges have to be rounded, and a round front has to be pu etc)
also, put most of the dimensions in order (there's a limit by the rules, for example the length has to be between 220-350 cm, so i put 350 max...stayed within the limits :-) ) but for some curved areas, i just relied on trial-and-error..still much to be done
and made it a solid coz i thought it d be easier to analyse

 

[xxChrisxx]

I realize this may be preliminary design, but ultimately do you have haccess to CFD or a windtunnel? A picture of the shape gives a general idea but conveys no useful information as to performance.

Cyrus' last post is dead right, you do need to have a clear idea of where you are going before even considering putting pen to paper for a proper drawing. Your initial drawing was fair enough to give us a general idea, but you've changed it now with seemingly no real reason.

Try and answer the following questions regarding the last picture.

Why have you changed the design?
Will that be better?
How do you know?

Unless you can give a comprehensive answer to all of those questions, don't change anything and don't draw anything because its meaningless if you don't know where you are heading.

 

[grey]

ok...well to be honest i did try to put in changes for reasons, tho i am not sure if they are good enough
1. this one is in AutoCAD, with (mostly) proper dimensions. It shows that such a design is possible while conforming to the limitations in dimensions set by the rules
(the angle at the 'trailing edge' is 20 degrees, but i think, given the low speeds, it should be ok; but to be sure, we need access to some CFD software (we don't have it at the moment, but i am sure some professor at the college would), like you say. Wind Tunnel i am afraid is not a possibility here.
2. Some changes i have made:
a) sweep-up at the underside at the rear

http://www.cardesignonline.com/design/aerodynamics/aerodynamics2.php][/PLAIN]
The underbody should be as smooth and continuous as possible, and should sweep up slightly at the rear

b) tried to smooth off the edges, not all of them yet though

http://www.cardesignonline.com/design/aerodynamics/aerodynamics2.php][/PLAIN]
There should be no sharp angles (except where it is necessary to avoid cross-wind instability).

How do i know? Thats where i am stuck...is there a way of telling without using softwares or wind-tunnels?

 

[PFMike]

Hi Grey
Good luck on your car building and racing efforts. I think you'll get much more focused answers on this forum if we have a better idea of what your goal is. I've competed and served as a judge in numerous car building and racing events and it always comes down to the goals and rules. You've mentioned "rules" a few times, what is the event that you want to compete in? What's the website for the rules? Depending on the event requirements I may be able to point you to some very useful resources that can help you develop your car.

 

[grey]

my apologies for this impertinence all along...
so i think i was missing this trick...really sorry...
number of passengers= 1 (min weight 70 kg)
weight of vehicle = 160 kg (max)
length = 220-350 cm
width = 120-130 cm
height = 100-130 cm
wheel base= 120 cm (min)
axle length = 100 cm (front, min), 80 cm (rear, min)
wheel rim dia (16-17 inches)

i think these are the main things
power source = there's a choice, but we have decided on IC engine with gasoline most prolly our fuel
hybrid is an option we are considering but that depends on the weight-power ratio, iff we can squeeze into the limits
obviously, i can't give you the power of the engine coz that i think should be worked out after we know what we are up against (in terms of how much to move, and drag)

the rest you can see in the rules (for UrbanConcept) in the site

the event is shell eco marathon asia
http://www.shell.com/home/content/eco-marathon-en/asia/welcome_sem_asia.html

obviously, i reread Cyrus' last post again...so much data regarding the constraints i have posted. regarding looking up research papers, i am not sure as to what to look for...design? our key aim is to reduce drag (my aim is to keep the co-efficient of drag x frontal area <0.1) and minimise weight (that obviously depends on materials, surface area, power system etc, which we can't get a figure until the design of the chassis is done)

if anything, i have missed, do tell so i post it