Aerodynamics for beginners

[AMMO]

I'm not sure I believe this business about flared wings making a Caterham front end lift at high speed. My car is very stable, but then it only does 110 mph on the clock. Probably closer to 100 mph in real life.

 

When I first got my car about ten months ago I did a little experiment. I tried with all my strength to lift the front end of the car by grabbing hold of one of the top wishbones. The car moved about 3 mm. I then hooked a fishing scale to a front flared wing and reached 8 kilos before the wing flexed and started making creaking sounds. The car itself didn't move. Surely to lift a few hundred kilos of front end you need something stronger than this?

 

Would you be able to lift an aeroplane by it's wings? I think yes. Maybe I'm wrong.

 

Please keep in mind that it's Saturday night I'm stuck indoors and there's nothing on the telly, otherwise I wouldn't be asking myself (or you) this.

 

AMMO

[R2D2]

I have flared wings on my 1989 car and I would have thought that if it did generate front end lift because of the wings there may be some damage to the aluminium panels by now and the rivnuts should be creasing the panels. I can't find any sign of damage,

[CHRIS CLARK]

When I used to marshal at Casle Coobe there was a V8 Westf1eld that circulated (towards the back of the pack) but he had cut 150 x 100mm square 'holes' out of the flared wing about two thirds back and covered it from underneath with mesh. I guessed this was to stop any presumed lift by bleeding off the air, but not allowing any large debris to come through.

 

Roger Swift until recently used a flared wing car to great effect on the circuits (but now has converted to the 'Dark Side').

 

Remember though; they look much better to start with thumbsup.gif

[Nigel Riches]

Dear Mr Ammo, when you attached the fishing scales to your front wing all the lifting force was felt at the point of attachment, hence lots of creaking noises as the fibreglass began to deform, this is not how things work in the real world, as the car progresses at greater speed so the pressure generated under the wings increases, this is caused by compression of air below the wing, but more accurately by de-pressurising the air over the upper surface, this is how a Boeing 747 wing can lift 360 tons of aeroplane, so although you maybe are travelling at some considerable speed, you wont actually take off, but the front wings will tend to generate "lift", if you are interested try sticking bits of woollen thread to the outside of the wings and go for a blat, see how the wool threads behave, whilst also keeping a wary eye out for Plod, as the pressure decreases the wool will lift away from the surface, may prove an interesting afternoons activity, you could post you findings here, so basically it's all down to aerodynamics, or lack there of.

Hope this is of some help. Nigel.

P.S. The most obvious draw back to "flared wings" is the "drag" ie lack of aerodynamics, as the opposite of "thrust" (acceleration)is "drag" (resistance to acceleration) and the opposite of "lift" is "weight", so it's all relative, if your in the aviation world.

[Miraz]

I'm not sure that it would be possible to detect any lift on the upper surface using woollen tell-tales, as they will give an indication of cirection of airflow - I would expect this to be laminar, or suffiently close to it that any surface turbulence would be difficult to spot from within the car.

 

One way of getting a rough approximation of the lift being generated would be to attempt to measure the difference in air pressure on each side of the wing surface and multiply by the surface area of the wing.

 

Miraz

[Rob Walker]

Caterham told me that they had done an experiment and slackened off the fixings on a set of flared wings and then run the car, they recon the wing was actually forced down and gives small down force not lift. Not shure if I believe this story

[AMMO]

Chris F, I agree.

 

Chris C, I think flared wings look nice too (but I don't dislike cycle wings either). Where I live flared wings are a neccessity. Lots of grit and rubbish on the roads. As it is the rear wings look like thay have been shot blasted. I hate to think what it would look like with cycle wings. There would hardly be any paint left.

 

Dear Mr. Nigelriches. You must have been reading my mind. I've been thinking about the wool tuft thing for some time. Could also help identify a good place to put air intake for induction system. I agree that flared wings probably limit forward motion. So the answer to being able to phisically lift a 747 by it's wings is?

 

Looking at Caterhams going down the front straight at Snetterton many cars had a slight nose up attitude. The cars front ends dipped briefly whilst changing gear and no power was at the driven wheels. So maybe weight transfer rather than aerodynamics is contributing more to the front end lift at speed. This may also be why the cars handle better with the rear higher than the front. This is the simplest form of anti-squat.

 

Also, the car has quite a bit of flat surface on the underside. It terminates at the rear in a slight up-sweep formed by the underside of the fuel tank. Could this be acting as a simple diffuser?

 

 

 

AMMO

[Blatman]

I have flares on one of the Westf1elds, and cycles on the other. Back to back, on the same, private, road (haven't managed a track yet) and at somewhere approaching max speed (130mph) I couldn't detect any front lift at all. Also, I believe that cycle wing cars generate significant drag caused by the rotating wheel being in close proximity to the wing. That is why Le Mans cars louvre the wings, to give the high pressure air somwhere to go. I would dispute the rear diffuser argument too, 'cos there are enough gaps at the rear for air to take the path of least resistance and flow up into the rear body area. I also have a "proper" rear diffuser on the cycle wing car, and I can feel no reduction in drag or rear end lift. The suspension and diff is nice and clean thoughwink.gif

[Slipper man]

Smeone (can't remember who, but it was someone who's opinions I respect) told me that the belief that flared wings creat lift is just an old wives tale. Apparently there is little measurable effect. They do create a little more drag (cd 0.64 v's 0.62), but the real advantage of cycle wings is just that they make the car much easier to work on.

 

Of yes, and they look much better....

[CHRIS CLARK]

Each to their own opinion Keith, and much as I respect your view to prefering the look of cyle wings...........YOU'RE WRONG !!!!!! Ha Ha

 

I would agree about the ease of working on the engine etc., but you said you don't do that anyway! Put a nice thick towel on the flared wing and you have the built in tool shelf.thumbsup.gif

Less dirt 'chucked' up too when used on our filthy roads.

[murph7355]

Chris - like your new "title" smile.gif

 

Cycle wings also let you see the wheels bobbing up and down. Which fascinates everyone who goes in the car for the first time.

 

They also make it look more like a racer (I originally wanted flared wings then after looking at plenty of cars went with cycles), and look significantly better with 13" wheels attached. And we all know the thing about 13" wheels...

[FH]

Love id=purple> the animation eek.gif

 

FH teeth.gif

[AMMO]

Miraz

 

I have a 3 ft. water manometer shaped as a U with an inlets at the top. I could then measure the pressure difference. Any ideas where to put the probes?

 

I have a very powerful flowbench which when set to exhaust chucks out something in the region of a 90 mph wind. If I made an adaptor with some ducting and blow air over the body work to do the wool tuft experiment. Maybe determine where to put the manometer probes. The duct could only be 5" in diameter. Am I wasting my time doing something like this?

 

An alternative is to drive the car with someone filming the wool tufts from another vehicle. Could try it out on the runway at Bentwters. Any volunteers (with a video camera) in the Suffolk area?

 

AMMO

[Miraz]

Ammo,

 

In order to get any meaningful numbers you would need to get airflow across the entire wing, so a 5" duct is not going to cut the mustard.

 

The manometer is probably too bulky to be practical, I'm sure that there must be some cheap, light digital pressure gauges available - possibly cheap desktop barometers???

 

As far as sites go it would be best to move the sensors around a bit in order to get an average figure, but in each site the inside/outside measurements should be taken at the same point on the wing.

 

The woollen tell-tales are not going to give any indication of the pressure (and hence the lift), although they will give an indication of the direction of airflow and the turbulence at that point.

 

My expectation is that you would find that there is negligible lift (pressure difference) generated by the flared wings, and that most of the increased drag is caused by the turbulent airflow in the underside of the rear portion of the wing.

 

Miraz

 

PS - Have a good look at the undercarriage on a commercial airliner (the 747 is a bad example) in order to work out if it is possible to pick it up by the wings.

[AMMO]

Miraz

 

Thanks for that. I'm pretty positive you can pick up a 747 by its wings. I'm also pretty positive you can't pick up a Caterham up by its front wings.

 

Steve Gilbert from Colchester, complete with video camera, will give me a hand with our psuedo-scientific experiment. Probably just a good excuse for a curry and a beer afterwards.

 

I'll post the findings, if any, sometime in the next few weeks.

 

AMMO

[EFA]

Ammo,

 

Make sure you go fast!

 

I chaged from flared wings to cycles in about 1994 in a 235bhp car which could hit 140mph through Fordwater at Goodwood. The front of the car went very light.

 

In the same car following a complete rebuild, with cycle wings, the car was much more stable.

 

Same applied on Lavant.

 

OK so the general setup of the car could make a difference, but this was an extreme difference.

 

I'd be interested on the outcome of your tests.

 

On the 747, the aircraft is predominatly lifted by the wings, which flex by some huge amount in rough contitions.

 

So although the front wings on a Seven don't feel robust enough to lift the front of the car, the load spread over quite a large area they probably do not move so much.

 

Also the compression of air running under the nosecone (the underside leading edge of which is shaped like a wing) will also contribute to the lift. When I made the CCC nosecones, we actually reshaped the bottom of the nosecone to reduce lift.

 

Finally I can confirm that the CCC small front wings do provide downforce. Not scientific this, but worth a laugh....

 

I fitted two new front wings to my car (the first of the wider variety) and having got fed up with the various methods of epoxy based attachment failing at high speed, I resorted to bolting them on with bolts which passed through the stays.

 

Now I was convinced that a lot of drag was caused by the air which gets trapped betweennthe tyre and the cycle wing, so by using some shortened wing stays I reduced thsi gap to around 20 - 25mm. I fixed the wings to the stays using 20mm mushroom allen headed anodised bolts (sexy!) and then went down the M3 for and hour to Lisa and Phil Lees house to pack upi the van for Le Mans. On arriving at Lisa's I noticed by ACB10's had some additional grooves in the tread. Although the ends of the bolts were 5mm+ away from the tyre surface static, at 80mph on the M3 they cut groves 3mm deep in the tyres!

 

So in order to conbfuse the issue more, perhaps flared wings create minimal downforce, and cycle wings create minimal+

 

Over to Mr Carmichael!

 

 

Fat Arn

The NOW PROVEN R500 Eaterid=red>

See the Lotus Seven Club 4 Counties Area Website hereid=green>

 

[Miraz]

An alternative theory....

 

ACB-10's are likely to balloon a little under centrifugal force at high speed, probably accounting for the contact with the bolts. You would need to exert a lot of downforce on the stays to achieve the same effect.

 

The gap between the cycle wing and tyre has to be large in order to create significant amounts of drag. The upper surface of the tyre will build up a zone of high-pressure under the wing, forcing the air over the top of the wing - the height of the zone is determined by a number of factors but the tread depth is the most significant. A reasonable rule of thumb would be about 6x the tread depth. The pressure zone is maintained by air forced forwards by the tyre surface and also sound radiating from the tread reflected in the underside of the wing.

 

It would be interesting to stick the bits of wool on the underside of the leading edge of the cycle wing to examine the airflow - I would not be surprised if the area immediately between the leading edge of the wing and the tyre was effectively static.

 

I think you will find that at 140mph everything starts to get very light even with the cycle wings, the increased feeling of stability is due to reduced turbulence and improved flow seperation not downforce.

 

The comparison with aircraft wing sections is not useful in understanding this issue, although it does make an interesting sideline. I've watched an A320 wing being flexed on a hydraulic test rig, which achieved over 7 meters of wingtip deflection from rest in each direction - this is hopefully more than an operational aircraft will ever experience. I believe that there was an incident with an Air China cargo 747 over Florida in the mid-eighties which involved an near-inverted power dive as a result of both pilots leaving the flight deck without correctly engaging the auto pilot. The aircraft was recovered and made a safe landing, but the airframe was written off - the wingtips at rest were nearly 4 meters higher than they should have been, and most of the tail had disintegrated.

 

Miraz

[Asklepios]

Aircraft can be lifted by their wings---Caterhams can not.The arguement about distribututing the weight of the spring balance would be disproven by rigging up a piece of wood to spread the load.I would bet a similar creaking would ensue. I think we can conclude it is unlikely that the wing actually acts as a wing to any extent.I suspect the airflow over the wing is turbulent and confused by the h/lamp.

The nose cone,if you take the lower face, is apparently an upside down "wing", and if we could evacuate the air under the car(or to the side),and ignoring the mirrored upper half of the nose,should we not be producing downforce?

[EFA]

Miraz,

 

In response to your comments I should add that I have GPS calibrated ACB10's when setting up the Stack unit on my car. The speed sensor gave linear output relative to the GPS all the way to 120mph, and when I reversed the calculations I got a tyre circumfernece figure 2mm smaller than Avons spec. My tires were not new. Therefore I don't think the tyres balloon.

 

The Caterham front cycle wings which were available at this time had a gap of around 3 inches under the wing at each end if you used VX race stays and 13" wheels. The wings I made are much smaller radius - 285mm vs 360mm for the Caterham item.

 

Fitted to the short stays the Caterham wings were definately an air brake.

 

The 140mph thing I have self conducted, and I don't think its down to my imagination. It was nervous before and is now rock solid. Others with EVO HPC's which started life with flared wings have reported the same!

 

 

 

 

 

 

 

Fat Arn

The NOW PROVEN R500 Eaterid=red>

See the Lotus Seven Club 4 Counties Area Website hereid=green>

 

[Miraz]

The distance recorded by your dash will be affected by the rolling radius at the point of contact, which would not be affected by ballooning of the unloaded portions of the tyre. The sidewall will flex out slightly at the point of contact as it carries the weight of the car, elsewhere the centrifugal load on the tyre causes the sidewall to flex inwards, this allows the surface of the tyre to balloon outwards.

 

All tyres behave this way to some extent, the ACB-10's are more prone to due to their thin sidewalls, low profiles and lightweight belts.

 

The alternative explanation is that the wings generated enough downforce to deflect the 10mm tube of the wing stay - this would require 40-60lbs of downforce on each side as minimum, given the surface area of the cycle wing this is extremely unlikely at the speeds you were doing. A well designed 6" aerofoil might produce this much force at that speed, but nothing else would.

 

I do take your point that a 3 inch gap would create noticeable drag, however the increased stability is more likely to be a result of smoother airflow, than of increased downforce. Any turbulence around the wing would feed back through the steering and make the car feel nervous, removing the turbulence would definately make the car feel more solid. A seven is a light weight car with the aerodynamics of a brick, at high speed it will be severely affected by any turbulence in the airflow around it, anything that can be done to reduce turbulence will noticeably improve high-speed handling.

 

I have some research papers around here somewhere that I contributed to whilst at uni that have some results of test performed on tyres under similar conditions - the research was aimed at modelling tyre behaviour on aircraft undercarriages, but the same principles apply.

 

Miraz M.Eng (Aero)!

 

 

[AMMO]

Two things.

 

Tyres definately balloon. I've built a lot of motorcycles in my time and on one occasion about 15 years ago managed to put extra grooves in my front tyre. I had made an ali fork brace to which the mudguard was fitted with bolts and nuts that had about 4 mm clearance between the tyre. At 90 mph on the motorway I could smell burnt rubber where the bolts had been rubbing. Spacing the fork brace cured the problem. Since then I always leave 10 / 12 mm.

 

Tyres do not get smaller in operation.

 

Tyres do not give 1:1 drive. There is a certain amount of slip especially at speed. Just because the wheel is turning at a speed that should allow you to do 140 mph when calculated doesn't mean you are doing that speed.

 

At Bruntigthorpe I had geared a bike to do 170 mph. It was hitting the rev limiter before it crossed the line but was doing only 164 /165 through the speed trap. Much scratching of heads. Until someone said "that's wheel slip mate".* Apparently drag-bikes slip their way all along the 1/4 mile. Dataloggers that are linked to wheel sensors are not a 100% accurate because of this.

 

Arnie, this may be why you thought your wheel was getting smaller rather than bigger.

 

* slipping between the the tyre and the road surface, not the tyre and the rim.

 

AMMO

 

Edited by - Ammo on 15 Oct 2001 08:25:56

[EFA]

Ammo, Miraz,

 

When I said 2mm smaller, this is because they were slightly worn hence the circumference would be smaller than norm.

 

Unless tyre slip is absolutely linear (say 3% across the entire speed range) and occurs at constant speed, I cannot agree with your theory. Sure a drag bike gets slip, but it accelerates at over 1G.

 

The GPS device measures speed between frequent plotted points tpo an accuracy of +/- 1m on a clear day with a gppd GPS device.

 

On the subject of ballooning I guess this could happen but only to an extent where it does not effect the rolling (as opposed to overall) radius of the tyre.

 

My stays are not a linear 10mm. They are half section 10mm around the 90 deg bend and down the section where the sidewall of the tyre is. This undoubtably eans they would flex more easily, but by how much?

 

I would not really think 40-60lbs of downforce is hard to create. The (albeit) front winglets of an F1 car must create around 500lbs from probably only 3x the surface area.

 

Bottom line however has to be then to reduce turbulence. Anyone got wind tunnel access big enough for my car?

 

 

 

Fat Arn

The NOW PROVEN R500 Eaterid=red>

See the Lotus Seven Club 4 Counties Area Website hereid=green>

 

[Peter Carmichael]

Arnie is talkign about a wheel sensor on an undriven wheel - no wheel slip.

[Miraz]

Tyre slip is not linear, especially on a high powered car like yours. The tyre has a limited coefficient of friction with the road surface. Because the friction is not infinite the tyre will slip under power and under braking. Your front-wheel sensor will not be affected by power slippage, but will under read under braking, and drag in normal driving. The slippage is exaggerated as imperfections in the road surface alter the amount of traction between the tyre and the road.

 

The reading from a wheel speed sensor is further distorted by changes in the tyre radius due to sidewall flex under changing load. It just isn't possible to get more than 95% accuracy by measuring the speed by which the wheels rotate - there are too many variable affecting the system.

 

GPS system are not accurate to +/- 1 meter - even using the higher grade military system. There is a detailed explanation of horizontal accuracy on commercial GPS systems here Typical RMS error rates on modern equipment are normally around 5 meters dependent on another whole range of variables including current satellite geometry, antennae characteristics, etc.... There are some very expensive (£100k+) aviation GPS units that use a secondary inertial reference, usually a ring-laser-gyro in order to achieve levels of accuracy around 1 meter.

 

The tyres on your car are not perfectly round, the contact patch is flattened by the weight of the car - the sidewalls and carcass flex in order to make this possible. Even at rest the radius measured to the road will be less than the radius measured to the top of the tyre. At speed the tyre will balloon on the unloaded portions of the carcass, but the loaded/contact portion of the carcass will not deform to the same extent as it is carrying the much larger gravity load of the car. It is probably worth pointing out that this starts to break down in extreme situations - have a look at the big crinkle wall slicks on a dragster as it starts a run as an example.

 

The front wing on an F1 car has about 4 times the leading edge width, and about 8 times the surface area, many designs also use multiple element and end-plates to improve efficiency. It is also a true aerofoil, the downforce is achieved by accelerating laminar airflow across the longer under side of wing. The cycle wing on a seven has turbulent high-pressure flow on the underside, and relatively low pressure flow above - even true aerofoil section would stall in these conditions.

 

You can find the results of some wind tunnel work here - the conclusion was that the standard cycle wings generated lift, blanking the space between the leading edge and the tyre reduced the lift - this is consistent with the pressure gradient described above. In order to achieve downforce you would need to increase the pressure above the wing until it was greater than the pressure below it.

 

Miraz

[Peter Carmichael]

Lotus are mucking around with moulded polycarbonate body shells for Elise variants, just like radio controlled car bodies and saving 30kg compared to the 340R. Who fancies a polycarbonate full width body for their Seven, leaving the Seven shape visible through it? You could even have a go at producing a scale model Panoz based on the Seven.

 

Admittedly, this approach rather rules out positive downforce as the bodywork would probably not be rigid enough to transfer significant forces. However, more neutral lift characteristics and reduced drag would be on the cards.

 

I made ~140mph (Stack speedo) down the straight at Snetterton on the club track day, still accelerating at 0.lG. Hmmm. I wonder what a top speed of 170mph feels like in a Seven.