- Literature Review
Internal combustion engines generate power from controlled explosions. In the case of a petrol engine these explosions are caused by sparkplugs igniting the fuel inside the engine's cylinders. Ideally the work produced from the combustion of fuel would all be converted into mechanical power to propel the car forward. However, due to the inefficiencies of an internal combustion engine a large amount of heat is also created. Some of this heat will exit the engine through the exhaust system but a large amount will be retained in the engine block and cylinder head (or heads in a V type engine). Removing the heat retained in the engine is the job of the cooling system. As a rule of thumb, ‘of all the heat produced from combustion, approximately ⅓ goes out the exhaust, ⅓ is dissipated through cooling and the remaining ⅓ is used to overcome friction and produce mechanical power' . It is important to note that this relationship applies at maximum power and that an even smaller percentage will go to the wheels when the engine is not making maximum power. While this is only a rule of thumb it does illustrate the importance of optimising the thermal efficiency of an engine.
The cooling system must remove heat from the engine in such a way as to maintain the optimum operating temperature regardless of any change in external conditions; such as wind speed across the radiator core and ambient temperature. If the cooling system dissipates too much heat then the engine will never reach its optimum temperature resulting in poor fuel economy and higher emissions. If the cooling system does not dissipate enough heat the engine temperature will keep rising until the engine self destructs.
- Formula SAE
Formula SAE is a class of racing where university students; conceive, design, build and race open-wheeled racing cars. In Australia the Formula SAE completion is held at the end of each year with teams from around Australia and the world making the trip to Melbourne to compete in a range of events to decide the champion. There are two types of event in the formula SAE competition, static events and dynamic events. The static events include; engineering design judging, cost report analysis and a presentation. Dynamic events include; acceleration, skid pan, autocross and endurance events. A further component of the endurance race is a fuel economy event. The restrictions placed on Formula SAE are limited in comparison with other forms of motorsport allowing students to use their ingenuity and imagination when designing components for their race cars. These relative freedoms in the regulations have resulted in different universities having different overall strategies for the design of their cars with no one strategy being more competitive than another.
The University of Newcastle has competed in Formula SAE since 2003, with 2004 being their most successful year to date. The team's strategy has been to design and build a totally new car every second year and to develop and tune the car in subsequent years; for example the 2006 car is an evolution of the 2005 car. 2007 saw a new design strategy for the team with a move away from the powerful Honda CBR600 4 cylinder engine, which had been used in the previous four cars, to the lighter more compact KTM 620cc single cylinder engine. The new compact engine allowed for the design of a smaller, lighter and more manoeuvrable car which is more suited to the autocross style of racing. The 2007 car was let down by a lack of testing which ultimately led to an early retirement from the event. However, despite the disappointing performance at the event the author believes the 2007 team made the right decision to design a smaller lighter car.
While the 2008 car is an evolution of the 2007 chassis the team decided to replace the KTM620 single cylinder with an Aprilia RXV550 V-Twin engine, involving the design of new intake and exhaust systems as well as the design of a new cooling system, which is the focus of this report. The 2008 Formula SAE team members and their chosen deign areas are listed in the following table.
- Literature Review
As previously mentioned, this study focuses on the design of an entire automotive cooling system. Although not much literature is available on the overall design of a cooling system, there has been a lot of papers published focusing on compact heat exchanger performance. Most of the heat exchangers in these studies have been characterised using either the Effectiveness-NTU or Log Mean Temperature Differential (LMTD) methods. Several papers have also been published looking at the effect of louvered fin geometry on the air side performance of a compact heat exchanger. The following section will discuss the books and papers that the author has found relevant to the current study.
In 1955 Kays and London wrote a book titled Compact Heat Exchangers , this book was based on data collected from two research programs .The first program was conducted in 1945 and was supported bythe U.S. Navy Bureau of Ships, the program was carried out at the U.S. Navy Engineering Experiment Station, Maryland USA. The second program was conducted in 1947 and was jointly sponsored by the Bureau of Aeronautics, the Office of Naval Research, the Bureau of Ships and the Atomic Energy Commission, the program was carried out at Stanford University. Cross checking was conducted between the two programs that yielded very consistent results. The test program continued after the publication of the book and the second edition contains data for an additional 25 samples bring the total number of test samples to over 90. The book is a great introduction to compact heat exchangers and contains a great chapter on heat exchanger design theory and pressure drop behaviour. However, with modern correlations for the air side performance of a heat exchanger the author of this study found no need to use the test data presented in this book.
Fundamentals of Heat and Mass Transfer , by Incropera, DeWitt, Bergman and Lavine, is a great introduction to heat transfer theory. The section of the book on compact heat exchangers only covers exchangers with round pipes and was based on the Kay and London text. However, the text contains a detailed section on extended surfaces (fins) and a comprehensive collection of data on the thermophysical properties of solids, liquids and gases.
Davenport reported in his paper of 1983, Correlation for heat transfer and flow friction characteristics of louvered fin, that the degree in which flow becomes louver directed is a function of the Reynolds number. At low Reynolds numbers the flow remained duct directed and at high Reynolds numbers the flow was louver directed, Davenport concluded that this was because the boundary layers on the louvers were thicker for slower air velocities and that these thicker boundary layers prevented the flow from becoming louver directed. Davenport also proposed a correlation for the Colburn factor j for louvered fins that formed triangular channels for 300<ReDh<4000, the correlation was based on 32 samples and he claimed that 95% of the experimental data had been correlated to within ±6%.
At the 9th International Heat Transfer Conference (1990) Sunden and Svantesson delivered a paper titled Thermal hydraulic performance of new multilouvered fins  that proposed a correlation for the Colburn j factor for heat exchangers with fins that form rectangular channels. The correlation was based on six sample cores and is valid for 100<ReLp<800. They concluded that a core with Fp/Lp = 1.52 and a louver angle of 39° would perform the best.
Chang and Wang's 1996 paper, Air side performance of brazed aluminium heat exchangers, developed a correlation, based on 27 sample cores, for the Colburn factor j for heat exchangers where the fins formed rectangular channels. Their correlation is valid for 100<ReLp<1000 and according to Chang and Wang 85% of the experimental data was correlated to within ± 10%.
In 1996 Chang and Wang also had another paper published, A generalized heat transfer correlation for louver fin geometry. In this paper they used their own data along with data from several other experiments to construct a general correlation for the Colburn factor j to describe all compact heat exchangers. In total 91 samples were used in the development of the correlation. Chang and Wang concluded that their generalized correlation describes 75% of the experimental data within ± 10%. In this paper Chang and Wang also examined how some existing correlations performed against their extensive data bank.
Chang and Wang developed a general friction correlation based the same 91 cores used in their earlier work. The correlation was presented in their 1999 paper titled A generalized friction correlation for louver fin geometry . They claimed that the correlation could predict 68.35% of experimental data to within ±10%. Due to inaccuracies in the correlation near ReLp=150 an amendment to the correlation was published in 2006 .
Olsson's 1996 paper, Heat transfer and pressure drop characteristics of ten radiator tubes , looked at the thermal and hydraulic performance of plane, dimpled, rib roughened and offset strip fin tubes. He determined the pressure drop and heat transfer characteristics for each tube for Reynolds numbers between 500 and 6000. He concluded that the rib roughened tubes showed the best results.
In 2000 Lin, Saunders and Watkins wrote a SAE Technical Paper titled The effect of changes in ambient and coolant radiator inlet temperature and coolant flowrate on specific dissipation . In this paper they developed a model for calculating specific dissipation. The ambient temperature was varied between 10°C and 50°C and the coolant radiator inlet temperature was varied between 60°C and 120°C. Their results indicated that ambient and coolant radiator inlet temperature had very little effect (less than 2%) on specific dissipation. However, the results showed that coolant flow rate had a significant effect on specific dissipation. Their results were verified using the Monash University wind tunnel.
Oliet, Oliva, Castro and Perez-Segarra, in their 2007 paper Parametric studies on automotive radiators , looked at the effect of air and coolant mass flow rates and air and coolant radiator inlet temperatures on radiator performance. They also looked at the effect of some geometrical radiator parameters on performance. They concluded that the fluid mass flow rate had a greater impact on performance than the fluid inlet temperatures.
Nuntaphan, Vithayasai, Kiatsiriroat and Wang studied the effect of the inclination angle of a radiator in their 2006 paper, Effect of inclination angle on free convection thermal performance of louver finned heat exchangers . Their results indicated that heat transfer performance generally dropped with rise in inclination angle. However, at angles between 30° and 45° there was a noticeable increase in performance.
- Objectives of This Study
The objective of this study is to design an effective and light weight cooling system for the Aprilia RXV550 V-twin engine used in The University of Newcastle 2008 Formula SAE racing car. The cooling system must control the temperature of the engine to ensure that the engine is always operating within its optimum temperature range. The cooling system must also be robust enough to function over a wide range of conditions.
A further objective is to study the effect of coolant flow rate, vehicle speed and ambient temperature on radiator performance.
- Water Cooling Systems
As mentioned before it is the job of the cooling system to remove heat from the engine. This is done by circulating liquid coolant through passages in the engine block and up through passages in the heads. As the coolant flows through the engine heat is transferred to the liquid via convection. The heated liquid then passes past the thermostat provided the engine is running at or above the optimum temperature and the thermostat is open. If the engine is not up to temperature and the thermostat is closed the coolant will bypass the radiator and be returned back to the engine until the engine reaches its optimum running temperature and the thermostat opens. Once the coolant passes the thermostat it is circulated through a hose to the radiator where the heat is removed from the liquid. The radiator is made up of a series of thin flattened tubes. As the coolant flows through the radiator heat is transferred to these tubes and then to the air flowing across the radiator core, this air flow is provided either by the velocity of vehicle, a thermatic fan or both. Once the heat has been removed from the coolant it is then circulated back to the engine ready to remove more heat. The circulation in a cooling system is driven by the water pump. Due to the high temperature the coolant is exposed to there is always a possibility of it boiling. By pressurising the cooling system the boiling point of the coolant is raised significantly. However, this high pressure might cause a clamp or gasket to fail resulting in a leak therefore it is necessary to control the cooling system pressure. Controlling the pressure is the job of the radiator cap.
The following figure was taken from the internet  and modified to better illustrate the scope of this project it shows the layout of a typical cooling system and the flow of coolant through the system.
The following section of the report outlines the purpose and operation of the individual cooling system components.
- The Radiator
Compact heat exchangers are used to dissipate heat when a large heat transfer surface area to per unit volume is required. There are many types of compact heat exchangers but the most suitable type for automotive water cooling is a tube and fin arrangement referred to as a radiator. Radiators are essentially a series of flattened tubes running between two tanks with fins extending between the tubes. Hot coolant from the engine flows into one of the radiator tanks, where the flow is divided up between the tubes, the coolant then flows through the tubes rejecting heat. When the coolant reaches the second tank the flow from the individual tubes are mixed and the coolant is returned to the engine ready to absorb more heat.
Radiator performance will be discussed in detail in a latter section of the report. However, the author believes it is important to give a brief overview in the introduction. Heat transfer from radiators can be broken up into three processes; convection from the coolant flowing through the radiator tubes, conduction through the tube walls and convection from the cold air flowing across the tubes. In a liquid-to-gas heat exchanger such as a radiator the air side convection offers the greatest resistance to heat transfer, this is because of the low heat capacity of gases. In radiators the air side heat transfer area is increased by using extended surfaces, or fins, this increase in area results in an increased heat dissipation rate. The denser the fins are packed the greater the surface area and generally the greater the heat dissipation rate per unit volume. However, if the fins are so dense that they start restricting the volume of air passing through the core the heat dissipation rate will be reduced. Another way of increasing air side heat transfer performance is through the use of louvered fins.
Modern radiators cores are made from aluminium with either aluminium or plastic tanks. Most Japanese and European cars have been fitted with aluminium radiators since the early 1980s. However, aluminium radiators were not used in Australian built cars until much later; with the Holden Commodore not having an aluminium radiator until 1989 and Ford Falcon not until 1998. Prior to aluminium, radiators were manufactured from brass tubes with copper fins soldered in place. The superior thermal conductance of aluminium is the main reason it has replaced copper and brass as the material of choice for compact heat exchangers. At 26.5°C the thermal conductivities of aluminium, copper, brass and lead (solder is approximately 70% lead) are 237W/m·K, 401 W/m·K, 110W/m.K and 35.3 W/m·K respectively. From these values it is evident that copper is a significantly better thermal conductor than aluminium which significantly better than brass. However, it is the solder used in copper/brass radiators that make them considerably less efficient than aluminium radiators.
- Water Pumps
It is the job of the water pump to circulate coolant around the cooling system; this is achieved by the pump producing a centrifugal force that draws coolant in from the radiator and sending it under pressure into the cooling jacket. The water pump also plays a role in controlling engine temperature. Water pumps are either driven mechanically or by a DC electric motor. The water pump must be powerful enough to overcome the friction of the water jacket and radiator and supply a sufficient flow rate for cooling.
Mechanically driven pumps can be driven from within the engine by the timing chain or externally via a pulley and serpentine belt. In both of these cases the pump speed will vary linearly with engine speed as the timing chain is connected to the cam (or cams) and the serpentine belt is driven by the crankshaft. This linear relationship is important in maintaining the engine at the optimum temperature. For example; say that an engine speeds up from 2500rpm to 4000rpm more power will be produced and therefore more heat must be removed from the engine but because the water pump speed increases by the same ratio the mass flow of the coolant in the system will also increase allowing more heat to be removed from the engine. Mechanical water pumps consist of; an impeller mounted on a drive shaft, pump housing and a gasket to seal the water pump to the engine block. Pulley driven pumps also require an additional seal to prevent coolant from leaking out around the spinning shaft. Electrical water pumps can either run at a constant speed or have their speed determined by an electronic controller.
- Thermostat and Radiator Bypass
The Aprilia RXV550 engine does not have a thermostat which is very rare for a water cooled engine. However, the author believes it is important to give a brief description of the operation of the thermostat as to better illustrate the challenge of designing a cooling system that does not incorporate a thermostat.
The thermostat is a valve used to control the temperature of an engine; the valve is controlled by the temperature of the coolant in the engine. If the coolant flowing through the engine is below the desired operating temperature the valve will block the flow of coolant to the radiator. Instead the coolant is circulated through a bypass system directly back to the engine, see Figure 1.2 A. This circulation of coolant through the engine ensures uniform heating of the coolant to avoid the creation of hot spots.
The valve action of the thermostat is caused by a sealed chamber containing a wax pellet. Below the desired operating temperature of the engine the wax pellet is solid, when the engine reaches its operating temperature the wax melts. As the wax melts it expands causing the sealed chamber to move, opening the valve. Once the valve is open the coolant will be directed to the radiator to remove heat, see Figure 1.2 B.
- Radiator Pressure Cap and Coolant Recovery System
As the temperature of a liquid increases it expands, because the cooling system is sealed this expansion results in an increase in pressure. This increase in pressure is desirable because the boiling point of a liquid increases with pressure. However, if the pressure becomes too great a gasket or seal may leak. A radiator cap will limit the pressure of the cooling system. Under normal operating conditions the lower gasket of the radiator cap will remain sealed to the filler neck, see Figure 1.3 A.
When the pressure of the cooling system rises to the limit that the radiator cap is designed for the pressure spring is compressed allowing a small volume of coolant to flow from the system into a reservoir tank, see Figure 1.3 B. The decrease in volume of coolant in the sealed system results in a decrease in pressure allowing the pressure spring to return to its original position, Figure 1.3 A.
When the engine cools, and the coolant volume contracts, a vacuum is formed in the system. This vacuum pressure causes the vacuum spring to extend opening the vacuum valve allowing coolant from the reservoir to return to the system, see Figure 1.3 C. Coolant is returned until the pressure reaches the design limit and the valve closes.
- Thermatic Fans
When a vehicle is moving slowly or is stationary with the engine running the air side convection may be insufficient to dissipate the heat produced by the engine causing the engine temperature to rise. Using a thermatic fan insures that there is always enough air flowing through the radiator to stop the engine overheating. In Formula SAE thermatic fans are extremely important because of the low average velocity (approximately 40km/h) of an autocross race.
Modern vehicles use electric fans that are controlled either by an electronic controller or the cars electronic control unit (ECU) and are programmed to only switch on when the engine reaches a certain temperature. In road cars where the radiator is mounted in front of the engine and space is limited the fan is usually mounted to the back of the radiator or just behind it resulting in non-uniform air flow through the radiator core. In racing cars that have the radiator mounted in a side pod it is best to mount the fan as far back from the radiator as practical, this results in a more uniform flow through the core and reduces the “blind spot” caused by hub of the fan.
Air, like all fluids, takes the path of least resistance so it is vital that the fan has a well designed shroud to force the air to flow through the radiator core and not around it. It is also vital that the cars bodywork also forces the air to flow through the radiator core