Exhaust Flow and Aero

[superford317]

It pains me to say this, as he has been one of my biggest hecklers, but “sparky” several posts ago made a comment about a discussion we were having about pulling a vacuum on the vehicle surface, he said “reverse that sucker and flush and blow” he may have been more correct than he knew at the time.
At the beginning of the last series of posts I made on riblets, I said please forgive me for the rather long posts, the same for this series of posts; this is a complicated subject that will need some explanations along the way, so here it goes.
Most of my previous posts dealt more with turbulent boundary layers and full bodied vehicles, most of this series of posts will deal with laminar boundary layers and stream liners.
The fastest speed a human has ever traveled was 24,791mph by Apollo 10 on its return trip from the moon on MAY 26 1969. There was no air to generate drag or friction heat on the surface, Outer space is very close to being a vacuum.
 The fastest “Publicly Acknowledged’ airplane ever built was the Lockheed  SR-71 Blackbird with a speed of MACH 3.2+, the SR-71 surface temperatures, due to aerodynamic heating, were around 800F at the nose, 1,300F on the engine cowlings and the cockpit canopy was over 550F. The Blackbirds JP-7 fuel was circulated behind the titanium outer skin on the wings leading edges and circulated around the electronics to help keep them cool. Asbestos was used in the hottest areas for protection also.   
A lot of research and millions of dollars have been spent on pulling suction on boundary layers, from the air force to NASA, small scale wind tunnel tests to full scale flying models implementing the process, but have been proven too complicated and the holes that the suction is pulled through on the vehicle surface have proven to be VERY hard to keep clean and operating properly. Blowing of air out of the vehicle surface through holes can be very beneficial also, as we shall see.
The method used and even the specific locations it should be employed will depend on the vehicle shape and the boundary layer, depending on if it is laminar or turbulent.
On a LSR streamliner with a laminar boundary layer, everything is nice, orderly and smooth and the car should be moving like a bat out of hades, but if you suddenly forced the boundary layer to detach it would be like popping a parachute and look like one also if you could see the airflow patterns. 
The boundary layer was first identified in 1904 by a young German physicist named Ludwig Prandtl, who was recognized later as the father of modern fluid mechanics, it was only a theory in a research paper, and it was many years later before it was actually proved to exist
You can take the exact same engine and depending on the aerodynamics of the vehicle it is placed in the vehicle can run 100MPH in one body design or run over 200MPH in another body design from nothing but aerodynamic changes in the body shape and surface texture.
Drag depends on velocity, as vehicle speed increases more power is required to overcome the drag. As speed is increased aerodynamic drag goes up quickly. Double the speed and you quadruple the drag.
Small gains in reducing aerodynamic drag can be as good as large gains in horsepower.
What is the boundary layer, you can’t really see it, smell it or taste it when it is attached to the vehicle in laminar or turbulent form, but when you lift from the accelerator you can really feel it is there from the deceleration caused by the aerodynamic drag forces and the pressure build up at the front of the vehicle.
Air has mass and there for the vehicle has to push it out of the way as it accelerates through it. The force generated from the air resisting being moved and the volume of air that is moved will depend largely on vehicle shape, vehicle speed and the air density. The air moving past the vehicle sticks to the surface slows down and starts forming layers that will vary in speed from zero to near zero speed at the surface of the vehicle to free stream airspeed several layers above the vehicle surface.
Each layer subsequently above the other moves a little faster than the one under it as you move above the vehicle surface until the outer most layers are at free stream speed. The layers between the vehicles surface and the top most layer just before reaching free stream speed is called the boundary layer. The boundary layer, as long as it is still attached can be considered as not much more than a thin film on the vehicle surface.
Viscosity is what can be considered as how sticky or gooey the air is and will be a factor in determining how thick the boundary layer will be. Pressure does not affect viscosity, at pressures of 7,000psi, it would only affect the air viscosity by about 10%; Temperature has by far the greatest effect on air viscosity. The height of the boundary layer will become smaller as viscosity is reduced. As the height of the boundary layer shrinks it becomes more stable. The velocity profile of the boundary layer can be greatly influenced by controlling the vehicle surface temperature, which in turn controls the viscosity. Viscosity in air is due to the transfer of energy between the different moving air layers, the energy being friction and heat, the layers in the boundary layer move at different speeds and so the air’s viscosity comes into play as stress builds in the different moving layers. Air exerts pressure forces and viscous forces on objects.
Viscosity in air is determined by the rate at which the air speed changes over distance and the energy that is transferred between the different layers.
As the temperature of a liquid increases its viscosity decreases, the air surrounding you is exactly the opposite. The viscosity of air increases as temperature increases.
Why the difference, as a liquid is heated the bond between the molecules is less so it tends to separate and become thinner. The bond holding the molecules together (cohesive) is greater than the energy transfer between the molecules (momentum).
In air, the bond between the molecules is a lot less while the movement and energy transfer between the molecules is a lot greater. As more heat is applied to the air the more the molecules move and the greater the energy transfer. The two are dominated by entirely different forces that react differently with energy input, in our case heating or cooling.

[superford317]

There are four states of matter, solid, liquid, gas and plasma, as you step up in matter, it contains a LOT more energy than the one preceding it. Think of it as hooking your air compressor to a plasma cutter and see what comes out. You are going from a lower state to a higher state. The same as going from the liquid to the air.   
The air moving along the vehicle surface will create shear stress on the surface, you can think of it as the stress forcing the air to rise up like a wave as the water approaches a beach, at the surface the air speed is at or near zero and as you move up in the height of the boundary layer the layer speeds increases. The lower layers are slower and the top layers are slightly faster, so the bottom layers are retarded and held back in there forward movement as the layers above move past the layers under them.
The height of the boundary layer can be influenced by how smooth the surface of the vehicle is, the air viscosity, air density, vehicle shape and the vehicle speed. 
As the boundary layer moves along the vehicles surface it will tend to grow taller as it moves farther back, the boundary layer is normally the thinnest at the front of the vehicle.
Air has very low viscosity so it does not want to stick to shapes very well as they move through it, air has less than 2% the viscosity of water. To keep the air attached and in laminar form where it will create the least drag along a vehicle body, a lot of restrictions will be placed on the body as to shape and design.
Taking the above mentioned streamliner with its very efficient body design as an example, the pressure gradient in the boundary layer will be distributed from front to back along the vehicle surface and the pressure in the boundary layer will decrease as the air flow moves from the front to the back of the vehicle, causing the boundary layer to stay attached and form a laminar boundary layer. If there is a rise in the boundary layer pressure in the direction of the airflow, caused by surface texture or body irregularities, the boundary layer will start to become turbulent and will totally separate from the vehicle surface if the pressure gradient becomes high enough, as I stated in an earlier post “THIS IS WHEN ALL OF THE BAD THINGS START TO HAPPEN”.
What really caused the boundary layer separation to happen though? The laminar boundary layer will follow the low pressure gradient. As was mentioned earlier, the speed of the air in the boundary layer is at or near zero at the surface of the vehicle and rises slowly as you climb up through the different layers until you reach free stream air speed at the top most layers. The slow moving air in the bottom layers at the vehicle surface has very little energy due to its slow speed and as the upper layers are moving faster they have more energy. As the boundary layer encountered the vehicle surface irregularities and the pressure in the boundary layer began to rise it will cause the slow moving bottom layers at the vehicle surface to slow more and actually stop and reverse flow and start forming vortices, the boundary layer goes from being laminar to turbulent at this point, the bottom layers in the boundary layer because of their low energy will have a high sensitivity to pressure variations, as the bottom layers slow down they become thicker, it would be like a rock in a moving stream stopping the flow of water, it will actually force the water striking the face to move down forming vortices and eddies and it will start to erode the base of the rock as the water flow is reversed and forced back up stream. A lot of people have found out about vortices and eddies when swimming.
The shear stress in the boundary layer at the vehicle surface will decrease as the square root of the distance from the area where the separation took place.     
The most effective way to promote laminar boundary layers and delay the transition to turbulence is with the initial vehicle body design that will promote stabile pressure along the length of the vehicle. A well designed streamliner has very graceful curves, no sharp bends or drop offs and gradual tapering body lines to encourage the air to stick to the vehicle surface, conform to its shape and promote laminar air flow and low drag. At the rear of the streamliner it should be gradually tapering so that the air is accelerated inwards to prevent a void at the rear and setting up turbulence. By minimizing the pressure gradients along the body, laminar air flow can be carried out over large distances on the vehicle surfaces, so that the friction levels are at laminar instead of turbulent levels, laminar flow generates much lower drag.
Corners and sharp bends cause a rapid deceleration of the airflow and can lead to boundary layer separation.
In certain conditions the boundary layer can become detached and actually reattach its self to the vehicle surface, when this happens it is called a laminar separation bubble.
Laminar to turbulent boundary layer transition can be controlled or delayed by influencing the speed of the boundary layer by altering the pressure gradient at the vehicle surface.
Boundary layer control can be split into two categories, active and passive control. The velocity of the boundary layer can be influenced by controlling the pressure gradient at the vehicles surface. The stability of the laminar boundary layer depends on the speed of the boundary layer near the vehicle surface. 
Turbulent boundary layers have more energy than laminar boundary layers and can handle higher pressure changes on the vehicle surface without becoming detached, but turbulent boundary layers produce much more friction drag than laminar layers, attached turbulent layers are still many times better than detached layers which cause many times more drag due to the ensuing pressure drag and the wake developing behind the vehicle, than turbulent boundary layers which are attached to the vehicle surface.
Active control measures require direct energy input to do there job and passive control requires no direct energy input for the work they do.
Examples of active boundary layer control systems are boundary layer heating, cooling, suction, blowing and even using flow driven vortex generators.
Examples of passive boundary layer control systems are fixed vortex generators such as the riblets, dimples V’s and vortex generating vanes.

[superford317]

The boundary layer control systems that are not controlled smartly from an external source will all have to be designed and tuned for there specific locations and the job they will perform on the vehicle surface, according to the boundary layer type and thickness they will be placed in.
The active boundary layer control systems with external controls can be used in a lot wider range of uses and conditions and can even be turned off when not required.
When steady vortices are introduced into separating boundary layers by passive means or steady blowing of air is used, active, they can postpone or prevent the boundary layer from separating from the vehicle surface. Turbulent flow can follow the vehicle shape easier than laminar flow.
The blowing of air into the boundary layer at an area close to separation reduces high pressure areas by accelerating the air in the boundary and can lead to reattachment of the flow.
Passive vortex generators transfer energy from the faster moving outer layers to the vehicles surface to delay boundary layer separation from the vehicle.
When an active system is used such as blowing air through small holes in the vehicle surface upstream of the separation location, it increases the mixing between the slow moving air at the vehicles surface and the faster moving outer boundary layers. When the air jets are randomly pulsed into the separating boundary layer, it is much more effective at delaying boundary layer separation than the steady blowing of air. Huge delays in the separation of the boundary layer can be gained with low levels of pulsating air flow into the boundary layer forming oscillating vortices.
The blowing of air into the boundary layer has a mixing effect and suction on the boundary layer through the surface of the vehicle has a calming effect on the boundary layer. The blowing of air through the vehicle surface will cause the boundary layer to become taller and suction will decrease the height of the boundary layer. A thin boundary layer is more stable than a thick boundary layer. Blowing air increases the speed of the boundary layer and the suction will decrease the boundary layer speed. Stresses in the boundary layer are increased by blowing and decreased by suction. With either blowing or suction, the effects on the boundary layer are greater on the downstream pressure gradients than at the spot where it is introduced into the boundary layer. Pressure fluctuations at the vehicle surface are regulated by speed differences in the whole boundary layer. The differing pressure gradients at the vehicle surface are related to turbulence and vortices that occur upstream. With blowing, skin friction will increase downstream because of the turbulence introduced to the boundary layer. Surface pressure will increase locally, immediately before and after the area where the air is blown and the pressure above the area where the blowing takes place will be less. Suction reduces turbulence at the vehicle surface. At the localized area where the suction is initiated, there will be lower pressure before and after the area and a higher pressure above the initiation point. Suction will have the exact opposite effects of blowing on the surface boundary layer. Downstream pressure gradients will greatly increase with surface blowing and be reduced with surface suction.
The boundary layer can be controlled and made to bend around small surface irregularities by using active vehicle surface heating and cooling. Removing heat from the vehicle surface boundary layer will have a stabilizing effect and adding heat into the boundary layer will have a destabilizing effect. Boundary layer transition from laminar to turbulent can be delayed by cooling the vehicle surface and can be brought about faster by adding heat to the vehicle surface. Upstream conditions will affect blowing more than suction, differing boundary layer speeds and pressures will have an influence on blowing more than it will with the suction.  An advantage to suction or blowing at the vehicle surface is it can be turned off and on at will. Surface suction is plagued with the problem of keeping the suction ports clear and clean, obviously blowing will not have this problem. 
A lot of control can be had over the boundary layer by the use of surface heating, cooling, suction or blowing. These forces can move the location of where the transition from laminar to turbulent flow takes place, control the thickness of the boundary layer, hasten or delay and even reverse the onset of laminar to turbulent flow.
The more turbulent the flow in the boundary layer the more effective these control systems are. On fully turbulent attached boundary layers a reduction in drag of 50% to 80% can be had. A vehicle with laminar boundary layers using surface cooling and low level spot suction could see drag reductions of up to 15%.
Using vehicle surface cooling in a laminar boundary layer will have another stabilizing effect in that it will promote a downward viscosity gradient in the boundary layer, the air will be less viscous as it comes closer to the vehicle surface, thereby reducing friction between the different boundary layers.  The opposite is true with surface heating as it causes the highest viscosity at the vehicle surface. Using surface cooling and suction a boundary layer can be bent around some pretty sharp turns.
Viscosity of the air is responsible for boundary layer separation; any control over the viscosity can have significant effects on boundary layer separation. Temperature has the greatest effect on air viscosity, lower the air temperature and decrease viscosity or increase the air temperature and increase viscosity.
If you are working on a streamliner or a full bodied vehicle with a turbulent boundary layer, surface cooling and suction will keep the boundary layer attached longer, adding a calming effect to the boundary layer, helping to reduce the amount of turbulence and encouraging it to stay attached over longer distances and reducing the height of the boundary layer also. The surface cooling will cause a downward viscosity gradient in the boundary layer, reducing skin friction. If there are areas where the boundary layer is about to become separated, due to a high pressure gradient from the vehicle surface, use blowing and surface heat just before the separation area to add more energy to the boundary layer.
Using full body blowing at low pressure is another alternative, blowing around 4psi at a vehicle speed of 190 MPH, caused a 30 percent reduction in surface friction of a turbulent boundary layer. The low pressure process can be used on either turbulent or laminar boundary layers. The low pressure blowing reduces surface roughness and surface speed of the boundary layer, with the benefit of lower surface friction. The low pressure air lifts the boundary layer above the vehicle surface allowing it to flow over the surface more freely and reduce the effect of surface roughness also. Think of it like an air hockey table. The boundary layer increasing in height as it goes from the front to the back of the vehicle will be reduced because of the lower surface friction, but the boundary layer height will initially be slightly greater due to the blowing effect.
By applying suction and blowing on a vehicle at the same time can be very helpful. You can apply blowing at the front and suction at the rear. To avoid over suction and creating more drag, the suction volume should be lees than the blowing volume. Blowing will reduce the vehicle surface friction drag and the suction will reduce the vehicle pressure drag. Low pressure surface blowing can be used where the surface pressures are high to reduce shear stress and the suction can be used in areas where the boundary layer is slowing and in danger of separation.  By using blowing with suction it is more effective than only suction by its self. When using suction combined with the blowing, if the blowing rate is higher than the suction, a thrust can be generated that will actually add to the vehicle acceleration.
Depending on the volume and pressure, blowing can be used to revive a slowing boundary layer in danger of separation or it can be used to raise the boundary layer ever so slightly and reduce friction drag at the vehicle surface.
It does not take very much heating, cooling, blowing or vacuum to influence a laminar boundary layer because it has such low momentum and energy.
By using surface cooling and suction it would reduce the height of the boundary layer and promote a more stable boundary layer that is less sensitive to surface irregularities and surface pressure gradients. By using active boundary layer control methods the streamliner body design can be more flexible in its design and shape also. The rear body should be tapering to the inside so the boundary layer is accelerated to the inside, creating low pressure gradients, that in turn help to keep the boundary layer attached and hopefully laminar for a longer time and distance.
Body design and boundary layer type will dictate the form of active or passive designs that will be used on the boundary layer. A particular vehicle may need surface cooling and suction at the front of the vehicle and may have to switch to blowing and surface heat towards the middle of the vehicle and then switching back to surface cooling and suction near the rear.
Since the boundary layer is so thin and low energy, which ever active system is used it does not take much energy to influence the boundary layer. Applications at the troubled areas are actually preferred over full body applications. If the blowing air pressure is too high it can literally blow the boundary layer off the vehicle and too much suction will increase surface friction due to pulling the boundary layer down against the vehicle surface too much. Localized suction will have up to 60% of the effectiveness compared to full body suction, but the full body suction will hold the boundary layer on for longer periods but cause more friction drag.
Actually moderation with any of the active controls is preferred. Pulsing blowing air is preferable over continuous blowing; low pressure is better than high pressure blowing and more effective.

[charlie101]

This is most interesting to read, Superford317! Bundle of horsepower seems to be possible to be carried in ice and batterys instead of inactive lead bars. I really hope you continue with a discussion on differences with a flat and a round belly and whats happening under the car at the close proximity to the ground. As usual most of my previous thoughts and belives are put on end and backward!

[superford317]

I am very happy that people are enjoying my posts and get a little more insight as to how and why things function as they do.
It took a considerable amount of my free time to make the lasts 2 posts as to riblets and the laminar boundary layer.
I started a while back on 2 more lengthy subjects I think will be eye opening and informative that everyone will enjoy, I still have a ways to go on them before they are ready. I have so many things I want to cover; it will take a while to get it done.
The one you were asking about, flat and round bottoms and there close proximity to the ground, I hope to do it a little later also.
It may be 3 or 4 weeks before I get to do very much, next week I have a meeting in Dubai UAE and then from there to the Philippines for a few weeks.

[Rex Schimmer]

Superford,
In your discussions you imply that the boundary layer on a streamliner can be laminar over the entire length of the body. I find this hard to believe especially with the present day cars that are running in the 400 mph area. Just looking at the Reynolds numbers for this type of vehicle would suggest that the transition from laminar to turbulent boundary layer may occur only a few feet from the nose of the car and that the majority of the car's body is subject to turbulent,( hopefully attached) air flow.

You do spend a lot of time discussing different ways to influence the boundary layer, heating, cooling, sucking and blowing but from a practical stand point related to the majority of land racing cars this is technology that is costly and very hard for the average builder to apply. I find it all very interesting but most not very practical for my build, but please continue to post as it is very interesting.

Rex

[Tman]

In the time you took to write all of that, Poteet and Main made 43 runs on the long course

[SPARKY]

317  hopefully I haven't caused you as much mental anguish as your detailed explanations will cause me before I get my pea sized brain to open up and down load this to understanding level.  Then to try to formulate an action plan to allow evaluation and assessment for several areas of concern for the skin package for my new lakester---

My working with laminar flow is limited to glider wings in the 60s-70s ---we cross hatched the gel coat on them with 320 grit on the leading edges to past the spar cap-- but that was primarily to expedite evaporation of rain droplets

how do overlapped skin joints affect the non laminar boundary flow an separated flow?

Thanks for challenging our pea sized AERO brains  

[Buickguy3]

  Interesting! I kept my "pea brain" focused [with the aid of some Bud], through the posts. There were times when I said: Damn! show me a picture. A little visual would help. The posts by 317 are very stimulating for an old guy's aging grey matter. Thank you for your well spent time.
   Doug 

[superford317]

I just made it back from a much deserved vacation, while I was looking over the south china sea, I started thinking and then started writing. i haven’t been idle, I made several articles I will post as time permits, I still need to refine and polish them up a little more. I enjoyed writing them and I think they will be very informative on many subjects to do with aerodynamics.
I hope everyone has kept “sparky” out of trouble since I have been away.

[superford317]

Please forgive the long post as this is a very complicated subject with many variables involved and I hope to explain it without a lot of big words, so that will require a lot of explanations along the way. I could easily make a book on the subject; I almost have here without trying.
I have been asked several times about vehicles with flat bottoms and rounded bottoms, the differences and what I thought about them.
This post covers flat bottom and rounded bottom lakesters in both free stream air and in close ground proximity with open wheels, full body vehicle underbody airflow, underbody tunnel design and downforce generation from the underbodies. I tried to make my thoughts as organized as I could, but it is still a bit jumbled.
There are 2 debates with LSR cars. Which are better, vehicles with flat undersides low to the ground and little airflow underneath or vehicles that are round or teardrop shape with more ground clearance and letting the airflow underneath. As with anything else related to LSR it will depend on many things, such as the speed of the vehicle, how much research and development will be done to the initial body design and the desired weight of the vehicle, the size of the vehicle and even the wheelbase of the vehicle. The teardrop shape with high ground clearance will have a little less drag, be less prone to getting air born in a spin, better directional stability and generate less down force. The rounded vehicle with a flat bottom ran close to the ground can generate a lot of down force if needed, be more prone to becoming air born in a spin and generate a little more drag.
The earliest known land vehicle with a streamlined body shape was developed by a Belgium race car driver and engineer Camille Jenatzy, he was nicknamed the “red devil” and set the land speed record 3 times. The car was raced in 1899 and was named the “never satisfied” and was the predecessor to all single seat race cars.
Aerodynamics is the study of the motion of air and how it interacts with objects. The things normally observed in aerodynamics are temperature, velocity, density and pressure. In the previous set of posts I discussed briefly the cohesive bond of the air molecules and how it is affected by temperature. Because air has viscosity it has resistance to being moved. The viscosity of air is less than 2%of the viscosity of water. Air resists the motion of a vehicle moving through it because of the cohesion between the air particles. The resistance is the force required to break the cohesion between the air particles and make room for the moving vehicle.
Air has volume, viscosity, weight and compressibility. Air consists of 77% nitrogen and 23% oxygen. Air is not considered compressed until its density has increased by 5%. Compressibility effects of the air are usually ignored at speeds less than 228MPH. Because air is compressible, its density will change with pressure, as pressure increases the weight of the air increases.
At sea level air is 800 times less dense than water. The viscosity of air is the single most important factor in the resistance of a moving vehicle through the air. Measurements of air are normally taken with dry air, with no humidity, because the presence of water vapor changes the density of the air. At sea level, air pressure is 14.7psi at 60’F, 1 pound of air occupies a volume of 13.1 cubic feet. As air temperature is reduced, the air pressures decreases and the air density increases.
The earth’s gravity exerts a force that keeps objects pulled to the ground. The viscosity of air helps to prevent movement through the air and the weight of the earth’s atmosphere exerts a pressure of 14.7psi at sea level and decreases with altitude. Air pressure and density decrease as altitude increases. At an altitude of about 19,000 feet air density and pressure is roughly reduced by 50%. You would think that with 50% less air that the aerodynamic drag would be much less and performance would be much greater, but at the same time there is only half as much air for the engine to use to produce power. That is why turbochargers are frequently used in airplanes to recover the lost engine performance. During WWII both sides used turbochargers, superchargers, nitrous oxide and water injection to try and gain an advantage over each other in air superiority.
By lowering the ride height, it lowers the center of gravity of the vehicle, thereby reducing weight transfer during acceleration on rear wheel drive vehicles. Rear traction under acceleration can be improved by increasing the vehicle ride height because rearward weight transfer will increase. By lowering the front of the vehicle and raising the rear, high speed stability and downforce will be increased because of the angle of the roof and the underbody.
The vehicles tire traction can be increases by adding more weight to the vehicle, thus increasing the downward pressure felt by the tire and increasing grip. The weight of the vehicle can be reduced, thereby increasing the rate of acceleration. The lost downward force on the tires can be replaced with aerodynamic down force, to create the same effective weight on the tires. At lower speeds the effective weight on the tires will be less, so the rate of acceleration can be increased. At higher speeds the down force will come on gradually and increase tire traction as the speed is increased.
Because vehicles operate close to the ground, the ground will influence the airflow underneath and around the vehicle. The ground will have a big effect on the vehicle performance, either good or bad depending on the vehicle design and the ground clearance.
Placing downforce generating devices or lift generating devices in close ground proximity will increase their effectiveness. Wings, vehicle underbodies or spoilers will have a greater effect and will actually aid each other in close ground proximity. The closer they are ran to the ground the greater there effect will be, unless the ground clearance is reduced so much that the boundary layer between the device and the ground touch or the pressure on the surface of the device becomes too great and causes the airflow to separate. The boundary layer from the vehicle contacting the ground boundary layer will cause the drag to increase dramatically. The 2 intermingling boundary layers will have a drag greater than the sum of the 2 separate boundary layers.
If rules do not permit a flat bottom, front air dam or the cleaning up of the underside roughness, it would be better to increase the vehicle ride height, in a situation like this, to reduce the lift generated and lower the drag. Rules permitting, the vehicle underside can have the floor rounded to the outside on each side near the door sills, so the high pressure air can be bled off to the outside. Large engine, transmission and differential coolers can be placed horizontally under the vehicle to clean up the

[superford317]

roughness. Large engine oil pans made very wide and long with an aerodynamic shape can be installed. The exhaust can have separate pipes, flattened and flared out as much as possible, ran side by side down the center of the vehicle underbody, to mask the underside roughness and lower the drag. 
The highest aerodynamic down force with the least amount of drag is generated at the vehicles underbody. The flat underbody is much simpler to produce and will generate large amounts of down force. The underbody with venturies built into it will generate larger amounts of down force. The flat underside with a front air inlet, side skirts and rear diffuser will mock a venture and allow the ground to serve as the bottom side of the venture, will be the best option. These Additions added to the flat floor to mock a venturi and increase down force, will be at a substantial cost in production time and research.
Daniel Bernoulli was a Swiss physicist and mathematician during the mid 1700’s. His experiments showed that velocity and pressure are related. Bernoulli’s work was very critical in designing aircraft in the twentieth century. Bernoulli’s equation helped in the design of the carburetor and explained how it functions.
Collin Chapman, working for team Lotus on F1 cars, designed the Lotus model 78 F1 car; the first race it won was the Long Beach GP race in 1978, Mario Andretti was the driver. The Lotus model 78 was the first ground effects car in racing and had a venturi shaped underbody. Venturi shaped undersides were banned in F1 in 1983, in an effort to try and reduce vehicle speeds. The venturi shaped floors were replaced with flat floors. By shaping the underside of the vehicle like a venturi the air under the vehicle can be accelerated there by lowering the pressure under the vehicle and creating down force. Even a small decrease in pressure over such a large area as the vehicle underside, can create substantial down force.
Drag force depends on vehicle velocity. Too much drag and you will be pulling a lot of unwanted air along with you that will be a hindrance to acceleration and top speed. In any form of racing top speed and acceleration are very important. To maximize acceleration and top speed you should minimize drag. Turbulent drag and friction drag do not increase at the same rate with increasing speed. Turbulent drag will increase at a much faster rate with increasing speed than frictional drag. When dealing with aerodynamics underneath a vehicle, there will be interference due to the ground being in close proximity and will to some extent, have an influence on the aerodynamic flow around the vehicle, it can be good or bad depending on if the vehicle underside is smooth or rough, the shape of the vehicle underside, the desired ground clearance and even the wheel base of the vehicle.
There will be a boundary layer generated on the underside of the vehicle as well as the ground, its effects can cause everything from generate down force to make the vehicle uncontrollable and become air born. A vehicle with a flat underside will have a minimum lift and drag at around 5in to 6in of ground clearance and downforce will start to be generated from there, as the ground clearance is reduced, and drag will gradually start to increase also. Under vehicle air velocity will increase as ground clearance is decreased. Velocity varies with area, if you reduce the area by 50% you will double the air velocity. At higher ground clearances the air flowing through the underbody will be affected by viscosity less, as the ground clearance is reduced it will be affected more by the viscosity of the air, as the speed of the airflow is increased. The most downforce will be generated from a flat floor, with a ground clearance of between 1.5in and 2.75in. As ground clearance is reduced, maximum downforce will gradually shift to the rear of the vehicle. The vehicle underside at a minimum should be looked at to reduce aerodynamic drag forces and to try and reduce its effects, to increase speed and stability by decreasing underbody pressure build up.
The use of airdams, side skirts and reducing the vehicle ground clearance can reduce the airflow underneath the vehicle, thereby reducing aerodynamic drag. Interference from the ground can cause pressure to build up underneath the vehicle and generate lift and in severe cases, cause the vehicle to become air born. 
I have never seen it in person, only in magazines and internet photos. Using speed demon as an example, it has side skirts to prevent air from entering under the body from the sides, has a smooth flat underside with a very gradual angle from front to rear to prevent pressure build up and the gradual upsweep at the rear of the vehicle will help to generate down force. A lot of time and effort went into preventing pressure build up under the vehicle, preventing lift and making the vehicle more stable at high speeds. This also encourages better airflow over the body by reducing the amount allowed under the body. At the very rear of the body, the floor making the very gradual up sweep will act as the diffuser and generate down force from the flat underbody. Due to its underbody design, speed demon generates down force from the underbody. If speed demon were to become unstable and spin, the underbody design will greatly contribute to pressure build up and be more likely to become air born. Rear down force on speed demon would be greatly increased if there were side panels on each side of the up swept floor at the rear of the vehicle extending down close to the ground. This would aid in the diffuser operation greatly, but at the same time cause the tendency to become air born in a spin to increase also.
The vehicle with the higher ground clearance and rounded aerodynamic shape will tend to be more neutral in down force and drag. The nose being higher in the air, will generate less down force because it forces less air up over the nose and across the upper body. This body design allows the air to go straight around the vehicle 360 degrees instead of having to bend around it, thereby reducing drag. The teardrop shape in open airflow has very good aerodynamic qualities, but when it is forced to run close to the ground, the airflow becomes constrained and behaves differently. The ground prevents the formation of symmetrical air flow and results in an increase in drag. By installing a flat bottom on the teardrop shape, its qualities of being close to the ground will be greatly enhanced and will counteract most of the negative characteristics.  If there were pressure trying to build up under the vehicle, the rounded bottom would bleed the high pressure air off to the outside. The rounded bottom with air flowing equally down both sides, top and bottom, would be like a rudder and aid in directional stability. The rounded bottom would generate very little if any lift, there is no floor area for the positive pressure to act against , if there were high pressure air under the vehicle it would bleed back into the free stream air due to the sharply rounded bottom . A flat bottom vehicle will have a smaller frontal area, than a rounded bottom vehicle. You can take 2 very aerodynamic shapes and mate them together and they can suddenly become very un-aerodynamic, because of the interference between the airflow patterns and boundary layers.
The best compromise of time money and effort would be a teardrop body design with its rounded and tapered shape to cut through the air. The vehicle should also have either a flat bottom or venturi shaped underside, both will generate down force, providing increased traction.
I like the design of the “spirit of sunshine” tank, teardrop body design with the flat floor close to the ground. With minimal horsepower, it has proven it’s self very capable of high speeds and good stability. It has one of the better body shapes with the engine exhaust at the most advantageous location. The vehicle with the flat bottom close to the ground, will usually generate a vortice off of the back corners of the vehicle where the rounded body meets the flat floor, increasing the drag slightly. The rounded vehicle with high ground clearance will usually not generate the rear vortices and have less drag.
The rounded vehicle with the flat bottom in close ground proximity will have a wider range of choices that can be used because of the design. The vehicle can be lowered to the ground, side skirts added and a front air inlet installed, to generate downforce from the underbody. The vehicle ground clearance can be increased to between 5in and 6in to generate little to no downforce and reduce the drag from the underbody.  At a cost to a little more drag but substantially more downforce, the air inlet can be allowed to let more airflow into the underbody region, add side skirts and a custom rear diffuser and increase the down force dramatically over only allowing air flow over the top of the vehicle body only.
To take full advantage of the benefits of either a flat floor or the installation of an airdam, side skirts will have to be installed, to block the air from the sides flowing into the newly created low pressure area underneath the vehicle. The closer the side skirts can be ran to the ground, the more downforce will be generated. With a side skirt gap of .800in, it could cause a loss of as much as 50% of the downforce from the underbody.
 A plain flat floor with a few degrees angle and an up sweep of the floor at the rear of between 5 degrees and 13 degrees will generate down force. The more elaborate you make the design the more down force will be created. Adding side skirts or center vanes to the rear up sweep area, which is the diffuser, will produce more down force. Changing the floor form being flat to a venturi shape will greatly increase the down force if the ground clearance is kept reasonably low.
A well designed venturi contoured underbody, vehicle side skirts and a low vehicle ground clearance will generate thousands of pounds of down force. A F1 car running under the constraints of a rule book the size of a new york phone book, dictating the underbody shape, size and locations, as well as the ride height and no side skirts, all in an effort to reduce down force and to reduce speeds, still generates around 600 pounds of down force from the underbody with a flat floor.
Where the venturi tunnel meets the rear diffuser, the lower edges should be left as sharp and squared off as possible, because there will be vortices created on each side of the diffuser, aiding the airflow and helping it to remain attached to the diffuser. The roof of the venturi, where the corners meet the wall, should have a large radius to prevent airflow separation. If you use a larger angle on the front air entrance or the rear diffuser to shift the center of pressure forward or backwards, it will be more sensitive to the main floor angle and ground clearance.         

[superford317]

Aerodynamic downforce generated from the underbody has the lowest drag to downforce ratio of any of the aerodynamic downforce generating devices.  Underbody downforce can be divided into active and passive.
Passive, being downforce generated when the vehicle is moving through the air and forcing the air to flow under or over the body to generate downforce.
Active, can be having a secondary power source pulling the air from under the vehicle, that is sealed with side skirts all the way around to the ground, causing a suction force to be generated under the vehicle. The active downforce will be irrelevant of speed, it will have the same downforce at rest as it will at 150MPH. If you reduced the pressure to -1psi on an area of 5 feet by 10 feet, you will generate over 6,000 pounds of downforce.
Texas native Jim Hall and Hap sharp founded Chaparral cars. Chaparral came from combining their last 2 names. Chaparral was the first race team to use fiberglass as a structural element and the first team to use scientifically designed airdams and spoilers. Chaparral built the car that inspired modern ground effects cars, the Chaparral 2J, also known as the vacuum cleaner. The Chaparral 2J had two 17in fans powered by a 45hp snowmobile engine and used plastic skirts to seal the underside of the car to the ground and pulled the air from under the car to generate downforce. It was outlawed by the sanctioning body after its first season of use. 8 years after the Chaparral 2J fan car, In F1, Brabham built the BT46B using side skirts to seal the underbody to the ground and a fan that was said, to be used to cool the engine, that pulled the air from under the vehicle, generating tremendous downforce, but was banned soon after by F1 also. If you reduced the pressure by only .18psi over a 5,000 square inch area it would generate 900 pounds of downforce. Not that it would matter to us, but the Chaparral 2J generated .4 to .5 more G force on a skid pad with the fans on than with them off. The closer the skirts hug the ground the more downforce will be generated.     
Flat floors or venturi contoured floors depend on mass airflow to generate downforce, the more airflow allowed in at the front air inlet and the more airflow ejected out the rear diffuser, the more downforce will be generated. Vehicles with airdams, side skirts and splitters rely on preventing air from going under a vehicle, reducing underbody drag and tire drag, thereby having a lower drag coefficient.
For vehicles that cannot be fitted with flat floors or venturi shaped floors there are several things that can be done to help them, rules permitting of course. The use of diffusers, airdams and splitters can be almost as effective as a flat floor, as far as down force and drag reduction are concerned, but will require more work and testing.
Oncoming air at the front of the vehicle goes through stagnation, slowing down and increasing in pressue.
An airdam fitted to the front of a vehicle, will wrap around the front of the vehicle and extend down close to the ground, reducing the size of the gap between the front of the vehicle and the ground. As the vehicle ground clearance is reduced with the airdam, downforce will increase and the drag will be reduced. At very small airdam to ground clearance, drag will begin to increase because of boundary layer interference and the pressure gradient becoming too high can actually stop the airflow under the airdam and cause the drag to increase and the downforce to decrease.
An airdam with very low ground clearance, will speed up the air flowing under it into the underbody, lowering the air pressure and creating some down force this way. The airdam reducing the airflow under the vehicle will reduce the drag generated by the tires, exhaust, underbody and frame. The benefits of the airdam are 2 fold, reduce the airflow going into the underbody thereby reducing underbody roughness and drag and to create a low pressure area thereby generating downforce.
The stagnation point at the front of the vehicle, where the air hits and builds up pressure, will be lowered closer to the ground when the airdam is fitted. With the fitting of the airdam and the lowering of the stagnation point, more air will be forced around the sides of the vehicle and more air will be forced over the top of the hood and at the same time less air will be allowed to pass into the vehicle underbody.
The air pressure on the hood will increase and at the same time the air pressure under the vehicle will be decreased, because of the airflow being changed. The difference in the pressure differential causes most of the added down force.
The negative pressure or suction on the vehicle underside will extend to about the middle of the vehicle, generating more down force on the underbody.
As down force is added at the front of the vehicle it will usually take away at the rear of the vehicle.
If a vehicle, that has already had a flat underside installed, has an airdam installed also, drag will usually be increased because the underside is already smooth, so there is nothing to create drag. The airdam its self creates drag, so total drag will usually be increased.
A splitter can be added to the airdam lower leading edge closest to the ground and will stick out horizontally towards the front of the vehicle. The splitter picks up down force from the high pressure stagnation point at the front of the vehicle that was just lowered closer to the ground with the addition of the airdam. The splitter should be between 3in and 6in in length and does not need to protrude further than the thickness of the stagnation area.
The low pressure area that previously existed under and behind the airdam, with the addition of the splitter, will generate even more down force than before, because the size of the low pressure area will be extended by adding the splitter, thus increasing the floor area for the low pressure to act on and the low pressure will be reduced even more also. The high pressure area above the splitter and the low pressure area under it will cause a large pressure differential that generates the down force. 
A diffuser can be added to the back of the splitter and generate even more down force from the airdam and splitter. The diffuser is an extension of the splitter, extending horizontality under the airdam and the back end of it being turned up. The splitter, airdam and diffuser form a simple venturi, the splitter extension being the throat and the rear upsweep being the diffuser, the rear upsweep will be the expanding cross section area for the airflow. The size of the splitter passing under the airdam and the size of the diffuser can be any size you want to make it, but the larger the size the more down force will be generated. The addition of the diffuser to the back of the splitter will lower the air pressure even more in the low pressure area under and behind the airdam, thereby increasing the down force even more.
On a vehicle with a flat floor and rear diffuser, the vehicle center of pressure can be moved forward or backwards by moving the location of the entrance of the vehicle floor to the diffuser. The highest downforce will be generated at the transition from the vehicle floor to the diffuser entrance. The diffuser entrance can be moved by changing the angle of the diffuser ramp. The angle of the diffuser floor can be between 5deg and 13deg, with an angle of 9deg to 10deg being most effective, diffuser angels over 14deg will cause the pressure to become too great and cause the airflow to separate, reducing downforce.
There will be vortices formed at the sides of the diffuser, that will improve the airflow through the diffuser.
The main job of the rear diffuser is to slow the speed of the under vehicle airflow and let the pressure rise to that of the external free stream airflow, before exiting into the free stream air. Some down force will be generated by the diffuser because normally the pressure in the diffuser will be lower than external pressure. Downforce is created not only in the diffuser but also under the entire floor area. The diffuser drives the airflow for the complete vehicle underbody. Because of the angle of the diffuser floor, its internal volume will increase as the aiflow moves to the back, causing the air from the underbody to expand as it passes through the diffuser, pulling air through the underbody. If the angle of the diffuser is too great, the airflow will separate, because it will not be able to overcome the pressure in the diffuser, causing pressure to build up under the vehicle. The angled flat floor will generate downforce by its self; the addition of a diffuser will increase the velocity of the air under the vehicle. The job of the diffuser is to convert the airflow’s kinetic energy or dynamic pressure into pressure rise or static pressure. The expansion of the air from the underbody in the diffuser slows the air, increasing the pressure. As the air is slowed down it is forced to become denser as the pressure increases. The diffuser can be a simple upsweep in the flat floor at the rear of the vehicle, but adding side plates that drop close to the ground and forming a tunnel will increase its effectiveness greatly. The side plates will allow the diffuser to generate more downforce at a lower speed. There will be a counter rotating vortice generated at the inside of the trailing edge of the diffuser side plates, enhancing airflow through the diffuser. The attached vortices inside the side plates will trail downstream into the wake behind the vehicle and cause a vortex induced suction. At very low ground clearances, the effect of the counter rotating vortices will be reduced as ground clearance is reduced. The counter rotating vortices will help to keep the arflow attached to the diffuser surface longer than it would be expected to at higher angles.   The diffuser angle should be between 9deg and 10deg to be most effective. The smooth flat floor of the vehicle leading to the diffuser will allow a larger area over which the low pressure air can act, creating more downforce. The diffuser itself does not create downforce, it is the area in front of the diffuser that the low pressure acts on that creates the downforce. Another benefit of the diffuser is that by it being located at the very rear of the vehicle, allows all of the mass airflow to help fill the void behind the vehicle, reducing the wake size and the resulting pressure drop and the induced pressure drag. The diffuser will reduce the turbulence in the wake, thereby decreasing the pressure drag.
At the lower angles for the diffuser, the downforce will be gradual as the ride height is reduced and at very low ground clearances the falloff of downforce will be more gradual also. At the higher angles for the diffuser, downforce will build more quickly and falloff quicker at reduced ground clearances.
It was becoming a lot more common, in F1 to dump the exhaust into the diffuser, so the hot expanding gasses can increase the effectiveness of the diffuser. As the vehicle is cornering and slowing down, the airflow will be reduced under the vehicle, lowering the downforce at the time when it is needed the most. At low vehicle speeds there will not be enough airflow under the vehicle to support the twin counter rotating vortices that help drive the underbody airflow through the diffuser and contribute to generating downforce. F1 race teams went to a lot of effort and research to dump the exhaust into the diffuser, the vehicle electronics were set up to keep the engine RPM’s up during cornering when the driver was off throttle and to dump more fuel into the exhaust so there would be more mass airflow generated to keep the vortice structures supported and generating more downforce in the underbody and diffuser at low speeds. Just in the last few weeks, F1 decided it will be banning the practice of dumping the exhaust into the diffuser for next year.
The downforce of the flat floor with a rear diffuser can be increased even more with the addition of a front air inlet and side skirts. The flat floor between the air inlet and the diffuser is where the low pressure acts and creates the largest majority of the downforce. The inlet and diffuser simply aid the airflow into and out of the underbody region. The leading edge of the inlet should have a radius to help with the airflow. Downforce will increase as the inlet angle is increased, because at the steeper inlet angles it lets the flat floor area be larger. The inlet angle should be between 5deg and 17deg. The shallower inlet angle will pull more of the surrounding air into the inlet, but at a sacrifice of floor area. With the addition of side skirts, front air inlet and a rear diffuser the flat floor can be made into an approximation of a venturi. 
The larger the floor area can be maintained, the more downforce will be created. Efficiency in the inlet and diffuser can be sacrificed somewhat, for the sake of a larger floor area, for the low pressure to act upon.       
A teardrop shape has the lowest drag in free air and as the ground clearance is reduced the drag will increase dramatically. The fineness ratio can be used to make an educated guess between teardrop shape bodies to determine which is more aerodynamic. The fineness ratio is the relation between the length to width. The ideal fineness ratio should be from 5 to 5.75. The teardrop shape should have the blunt nose forward and the thin end to the rear. The area where the greatest thickness should be is recommended to be about 0.25 to 0.30 percent of the length of the body from the front end. As people like to say “it’s not how you open the hole in the air, it’s how you close it” this being said, the body should gradually tapper from the thickest point to a point at the rear, this area being the most important from an aerodynamic standpoint. The airflow being accelerated to the inside at the rear will encourage the air to remain attached to the body as long as the pressure does not become too great and force the

[superford317]

airflow to become detached. A large portion of the blunt nose at the front can actually be cut off flat without greatly harming aerodynamics and drag, but irregularities or too steep of an angle on the rear of the teardrop shape can greatly increase drag, due to boundary layer separation.
A torpedo shape with a gradual bulge in the top and bottom at a ground clearance of 6in to 7in will have 50% more drag than free stream, if the ground clearance approaches near zero, the same shape will have the drag increase 500%. By making the bottom flat the drag can be reduced dramatically at low ground clearances.
In the early 1980”s professor A. Morelli of the University of Italy, proved a streamlined body in close ground proximity, can have the same drag coefficient as a streamlined body in free air. It was called the morelli body and based on frontal area had a drag coefficient of 0.05.
The debate has been going on for years and will likely still be going on long after I am dead and gone. Which is better, flat and low VS round and tall will depend on the budget of the vehicle build, the amount of time one wish’s to invest, desired speed, available horsepower, if downforce is desired from the aerodynamics, suspension and even the surface being ran on. Considering there are several fiberglass tanks radially available, there is a good design that saves a lot of time and effort with construction, a flat bottom can be added to one if desired also. The vehicle with the lowest drag, which manages the airflow most efficiently, is the rounded body with the taller ground clearance, so it would be better for vehicles that are power limited. The flat bottom vehicle close to the ground will generate a lot of downforce if desired, have slightly more drag and be more prone to become airborn in a spin situation.
If extra ballast weight is carried to aid traction, you can choose to generate downforce with the underbody and carry less weight in the vehicle. The vehicle being lighter will have a faster acceleration rate. The downforce will be gradual and increase with vehicle speed, producing a linear downforce and traction curve. The lighter vehicle will produce less wear and tear on parts due to its lighter weight. Racing on a very rough surface will generate fluctuations in the downforce and traction, as the body makes undulations over the bumps. Spring rate will have to be calculated to account for the added weight of the aerodynamic downforce at high speed.
It has been proved in wind tunnels and at the track, vehicles can be designed with close to the same drag weather they are running in close ground proximity or in free stream air. In the end it will depend on if you want to generate downforce with the shape, have more confidence in a spin of not becoming airborn or the ease of construction. 
As an inverted wing close to the ground has its ground clearance reduced, its downforce generation properties are magnified, as the ground clearance decreases, until it is very close to the ground then drag increases and downforce falls off quickly. A regular lifting airfoil will have its properties increased as ground clearance decreases. The same for any aerodynamic device, diffusers, flat vehicle undersides for downforce, splitters and airdams, all work better in reduced ground clearance or ground effect as it has become known.
Whatever tendencies a vehicle has built into its underside will be increased as its ground clearance is reduced. Most full bodied modern vehicles will generate between 140lbs and 180lbs of lift due to its underside roughness and shape and design of the body. If nothing is changed, but the ground clearance is reduced, the amount of lift generated, will increase as ground clearance is reduced.
A vehicle with nothing but a flat smooth underside installed can be neutral, generate lift or generate downforce, depending on the angle of the underside and the effect will be magnified with reduced ground clearance.
Everything being equal though, as ground clearance is reduced, drag will increase, because of the viscous effects of the airflow under the vehicle will be increased as ground clearance is reduced. As ground clearances become smaller air velocity will increase, thus increasing its viscous drag. As the ground clearances fall below 5in to 6in, the effects of the ground will begin to be felt and be magnified as the ground clearance is reduced.   
 

[Peter Jack]

Superford, you sometimes give me a headache with all the reading but I have to admit that I appreciate your interpreting and writing all this stuff out instead of plagiarizing from others and using cut and paste. We've had those experiences on this website in the past so a lot of us are a little gun shy when we see these types of posts.

Pete