Leaving ill-informed opinion aside and concentrating on the mechanical and physical aspects of titanium alloys, it is still a strange material. Even though the alloys used for motorsport components are relatively strong, their behaviour when subjected to even modest levels of surface stress is extremely poor. For example, it is impossible to use titanium in almost any application where sliding is involved. Despite its strength when loaded in tension or bending, we cannot use titanium or its alloys for sliding applications unless we take measures to improve the surface through the use of hard engineering coatings such as chromium nitride (CrN), titanium nitride (TiN), diamond-like carbon (DLC) or a metallic coating such as molybdenum.

Titanium also has notably low thermal conductivity, and its ability to insulate components from the effects of heat gives it some applications in engines, where the material is allowed. Titanium, in being viewed (wrongly, in my opinion) as an exotic material, is banned in many governing bodies’ regulations for race engines. It can be used to provide an insulating ‘gasket’ for some applications, and titanium fasteners reduce heat transfer through joints compared to steel and other materials. Even though we might see the fasteners as representing a small cross-sectional area compared to the rest of the joint, the heat transfer can be significant, and measurable reductions in component temperatures can be measured by changing from steel fasteners to titanium.

The combination of thermal conductivity, density and specific heat capacity mean titanium has a very low thermal diffusivity, which is the measure of a material’s ability to conduct heat divided by its ability to store heat. Materials with low thermal diffusivity show high temperature gradients in response to local heating, as the transfer of heat from ‘hotspots’ to cooler areas is poor. The thermal diffusivity of titanium is less than 50% that of steel, and about 10% that of aluminium.

Titanium, while unremarkable in terms of strength and stiffness, has a low ratio of stiffness to strength. In some circumstances, notably in the case of fasteners, this is a very attractive combination. For a given service load on a joint, a reduction in fastener stiffness leads to reduction in cyclic load borne by the fastener. This is why, in highly stressed fasteners such as cylinder head studs or con rod bolts, the shanks are waisted. In some cases, changing from steel fasteners to titanium can also be accompanied by a reduction in the dimensions of the fasteners without a loss of durability.

The density of titanium compared to steel means that, for an equivalent component mass, more titanium can be used. For a component such as a con rod, this represents an opportunity to improve the bending and torsional stiffness. The actual deflection in normal operation may not be a concern, but stiffness has an important influence on vibration and resonance, and the increased stiffness of a titanium component may be significant in this respect.

Written by Wayne Ward

The applications for which these tool steels were developed means there is a range of materials with the ability to resist cracking or loss of strength at moderately elevated temperatures, or steels that can be hardened in air rather than oil or water, and by which method they can minimise distortion. Most tool steels are designed to be resistant to wear from a number of mechanisms – cutting tools need to resist abrasive and adhesive wear under intense sliding conditions with poor lubrication, while mould tools need to resist wear under the action of viscous, hot fluids flowing over their surfaces.

The combination of wear resistance and high fatigue strength, along with a degree of temperature resistance, makes some tools steels attractive candidates for manufacturing race engine components. This is often not a trivial undertaking, as the highly alloyed nature of these steels means the heat treatment processes involved are very specialised. Many tool steels require double and even triple tempering steps in order to gain the full benefit of their enhanced properties, and some steels also incorporate cryogenic treatments to fully transform their structures. The benefits of using the materials therefore needs to be substantial.

Tool steel camshafts are specified where a combination of high stress and endurance is required. Some also make good candidate materials for camshafts coated with DLC, as they do not to soften at the temperatures used during the coating process. As we might expect, tool steels are also sometimes used for cam followers.

Continuing with valvetrains, several suppliers specify tool steels for pushrods used in overhead valve race engines, while others who use tool steel ends fitted to pushrods produced from more conventional steels.

Gears for pump and camshaft drives are sometimes also made from tool steels. Traditionally these gears are made from carburised steels, but the carburising process (also known as case hardening) often has the disadvantage of introducing significant distortion. There is often very little difference in cost between tool steel and carburised gears made in small batches.

Tool steels are sometimes used for piston pins, and again we see that the materials they replace are often those where carburising or nitriding has been used previously.

A number of tool steels are used where high-strength fasteners are used, and con rod bolts are sometimes specified in tool steels such as H13.

Some tool steels, especially powder metallurgy types, can be so highly alloyed that their elastic moduli are significantly increased relative to traditional steels, combined with a reduced density. Several of them show a 10% increase in modulus, and I have seen at least one which has a specific modulus (elastic modulus divided by density) that is 20% higher than a typical steel. Increased stiffness is very attractive to design and development engineers, especially where valvetrain and cranktrain components are concerned.

Written by Wayne Ward

However, because motorsport’s use of MMCs has only ever been a very small niche compared to other markets such as automotive development of brakes, aerospace and defence and so on, the development of MMC materials has not been affected to any great extent by our enforced lack of interest.

Many researchers in the field of MMC development are concentrating on metal-matrix nano-composites, sometimes referred to as MMnC materials. The reinforcement in MMCs can take one of three forms – continuous fibres, whiskers/short fibres or particulates.

The clue to the thrust of MMnC development activities lies in the name. The reinforcement particles in nano-composite materials are very much smaller than has customarily been used, and the results of doing so seem promising. The particulate size is not necessarily in the realm of single-digit nanometres, but the scales are much smaller than the so-called micro-scale composites, and there is certainly some work being done where nano-scale reinforcements are below 10 nanometres in size. However, people working on nano-composites are generally discussing MMCs which have particulate reinforcement that can be several hundred nanometres in size rather than multiple micrometres.

We hear a great deal about the possible applications of carbon nanotubes and how they might bring about a revolution in materials, and there are a lot of researchers working on carbon nanotube-reinforced materials, particularly based on aluminium and titanium matrices. Such materials are often denoted by the letters CNT, so for example you will find a number of technical papers by searching for ‘Al-CNT’. CNT materials look extremely promising, but it remains to be seen whether the much-vaunted step change in materials properties actually happens.

However, much of the research into metal-matrix nano-composites is based on smaller-scale reinforcement using ‘traditional’ MMC fillers such as titanium diboride (TiB₂), aluminium oxide (Al₂O₃) or silicon carbide (SiC), and these too are yielding improvements in mechanical properties such as strength and stiffness.  

There are some processing difficulties in dealing with very fine particulate reinforcements, such as the particulates not being dispersed evenly within the matrix and poor bonding between the matrix and reinforcement. Consequently there is a great deal of research being done on the processing of the materials. One reason that nano-composite materials aren’t more widely available is that the optimum processing methods and associated parameters have not yet been found. There are a number of liquid-state techniques, where the material is processed in its molten state, which show promise, but these often require some expensive secondary processing to achieve optimum properties.

Powder metallurgy methods are perhaps the closest to being perfected for commercial use. Here, the matrix and reinforcement are thoroughly mixed as finely divided powdered solids before being sintered under pressure, and there are commercially (though not widely) available aluminium nano-composites. Such materials have been used for engine components, and they offer distinct advantages for customers, not only in terms of improved properties but in the ease of manufacturing. Where micro-scale aluminium MMCs often required special cutting tools, aluminium nano-composites can be machined conventionally.

Written by Wayne Ward

The reason that aluminium is so widely used is its combination of properties and its cost. However, people have looked seriously at magnesium in the past as a piston material. While they have found it wanting when compared to aluminium, it is not beyond reason to expect people to look again in the light of new racing regulations and recent developments in engine and manufacturing technologies.

So when aluminium is such a well understood and relatively cheap material, why would people look towards magnesium as a possible replacement? Well, it’s a question of density – at 1800 kg per cubic metre, aluminium is 50% more dense – but the stiffness of magnesium compared to aluminium is pretty much proportional to their densities, (that is, magnesium is about two-thirds of the stiffness of aluminium). With most of the piston’s stiffness being a function of its behaviour in bending, the required increase in section thickness to achieve the same stiffness as aluminium would still leave the magnesium piston much lighter. It is true that the room-temperature strength of good wrought magnesium alloys is lower than typical aluminium piston alloys, but at higher temperatures magnesium stands up to much closer scrutiny. For example, at 250 C the tensile strength of WE43 magnesium is higher than that of 2618, which is a favoured aluminium piston alloy.

With the combination of high-temperature strength, sufficient stiffness and very low density, shouldn’t magnesium be a popular piston material? The most important reason why we don’t use it is its low thermal conductivity – it has around 60% lower thermal conductivity than 2618, which means that without considerable effort put into cooling, the magnesium piston will tend to run much hotter than its aluminium counterpart. This degrades performance, as some of the heat in the piston will be rejected to the incoming fresh charge, raising its temperature and lowering volumetric efficiency in the process, although there ought to be a significant decrease in the heat rejected from the combustion process, so we could expect fuel conversion efficiency to be improved.

Where a set of racing rules rewards increasing power from a given fixed engine capacity, especially with an imposed maximum engine speed, any loss in volumetric efficiency is going to hurt performance. However, where fuel efficiency is at a premium – and this is an important factor in many forms of racing – improved fuel conversion efficiency is to be highly valued. There are types of racing where engine size, engine speed and breathing capacity are free, but the instantaneous fuel flow rate and total fuel load for the race are limited. In these types of racing, where the engine is constricted in terms of performance by an upper fuel flow limit, then fuel conversion efficiency equals performance. With the fashion for regulation that rewards efficiency, the magnesium piston may yet see a new dawn.

Written by Wayne Ward

However, for engineers concerned with race engines, hybrid systems and transmissions, the most eagerly watched group of materials are the metals. RP metallic materials have been available for a more than a decade – I had my first laser-sintered parts made about ten years ago for an engine test rig – although their visual quality or structural integrity often left much to be desired. Now though, the range of materials has improved, as has the quality of the powdered material. During the past few years, the machinery which produces the parts has also become more capable and more affordable.

The range of metallic materials covers everything from aluminium to high-temperature superalloys. Inconel is a case in point, which is currently used for a lot of development components. Inconel 718 is used in sheet-metal form for exhaust systems, so its use in making RP exhaust components should come as no surprise. The ability to make complex designs that would be impossible to weld gives design and development engineers more avenues for investigation.

The technique of laser sintering allows us to produce complex internal structures in components, the only limiting factor being the need to consider the removal of unconsolidated powder from the component. I have seen a sectioned Inconel poppet valve with a complex ‘matrix’ inside, and there is work being done to optimise material properties for this and similar applications with a view to it becoming a viable mass-production process in years to come.

The disadvantage with the use of high-temperature materials for valve manufacture is high component mass. Where boosted engines are concerned though, especially those that have to last more than a few hundred kilometres, the mass penalty is something that has to be accepted. The hollow poppet valve with a load-bearing internal matrix is a tempting prospect, and may allow the use of Inconel valves with a much smaller weight disadvantage compared to titanium valves than has traditionally been the case.

There is one piston manufacturer who has an almost skeletal prototype Inconel piston in its display cabinet. There is no evidence to suggest that this component has ever been run in an engine, but it does show some of the design possibilities for a piston developed for high-temperature use. This idea is discussed in more detail in issue 78 (May 2014) of Race Engine Technology.

Written by Wayne Ward

Aluminium cylinder liners are very popular, as they conduct heat well and the similar coefficient of expansion between the liner and piston means the cold clearances can be closer to ideal than with a steel liner. Compared to steels though, the allowable stresses are low, so the material’s thickness increases in order to keep stress within allowable limits. That means the water jacket and the material surrounding it are larger, and the other obvious implication is that, for a given bore size, the bore centres are further apart than would be the case for steel.

The obvious effects are that the engine block is heavier, and the crankshaft is longer and heavier. Less obvious effects are that the cylinder head fasteners are forced further from the combustion chamber, increasing bending stresses. Despite the aluminium cylinder liners themselves usually being lighter than their steel counterparts, it is often the case that, for a new engine design of given bore and stroke, with aluminium liners the engine is heavier overall.

Aluminium liners, where they are used in water-cooled blocks and in direct contact with cooling water, can also suffer from cavitation damage owing to the low strength of the material, and it can be necessary to put a coating on the exterior of the liner to make the surface more resistant to cavitation. Steel liners don’t suffer from this problem, although they may require surface treatments to prevent excessive corrosion.

The thermal conductivity of a cylinder liner isn’t as bad as the comparison of the coefficients of thermal conductivity would first appear. The fact that a steel cylinder liner can be much thinner than its aluminium counterpart to match either displacement or fatigue stresses means that the rate of heat transfer can be almost as high for a steel component as for an aluminium part.

However, there is nothing about the strength of steel that can help offset the fact that the coefficient of thermal expansion is much lower than that of aluminium. Where a ‘standard’ grade of steel is used, we simply have to accept that we need a much larger cold clearance between liner and piston; there is at least one grade of steel though that has been developed specifically to match the coefficient of thermal expansion of aluminium. Although not developed for use as a cylinder liner material, it has been used for cylinder liners. The material is not widely available, nor is it easily sourced in suitable sizes for liners. 

The ‘problem’ with the grade of high-expansion steel I am familiar with is that it does not respond to thermal strengthening, although it does work-harden. However, it is not easy to work-harden the large bar or tube sizes that are required for liners, and the low strength of the non-work hardened material may be too low for many applications.

Written by Wayne Ward

There are several limits on increasing engine speed, but one of those against which engineers continually find themselves battling is the valvetrain, and in particular being able to keep the valvetrain under control. In order to do this, engineers generally like to make the system as stiff as possible between the cam lobe and the valve head. Reduction of reciprocating mass is also extremely important if we are to prevent parts of the valvetrain becoming separated from each other as the engine speed increases.

In seeking to minimise valvetrain reciprocating mass, aluminium is an attractive material. If we look first at the common valvetrain reciprocating components between overhead valve and overhead cam engines – namely the valve, spring, retainer, collets and lash cap – only retainers are regularly produced from aluminium, and these need to be examined regularly. Their bad reputation is mainly a result of people overstressing them or failing to examine them for signs of wear or fatigue. Aluminium metal matrix composites are also used for this purpose. The weight saving achieved through using aluminium retainers is certainly a useful proportion of reciprocating valvetrain mass.

Aluminium collets are sometimes used in race engines, but these offer very little weight-saving potential compared to larger components such as retainers.

Aluminium inlet valves have undergone trials in race engines, and some automotive car manufacturers are very interested in developing these for use.

For overhead cam engines with inverted bucket followers, aluminium has been used with some success by some car manufacturers, but such followers require a ceramic or steel shim against which the cam lobe acts, and can also place a limit on the maximum valve lift velocity that can be achieved. The maximum lift velocity is directly proportional to the maximum eccentricity of the cam-to-follower contact from its zero-lift point. The shim used with aluminium followers requires a ‘wall’ around it to retain it, and so the contact surface is appreciably smaller in diameter than the follower. Since the limit here is often the inability to fit a larger follower owing to the space between adjacent valves, using an aluminium follower with a separate shim can be a disadvantage.

In overhead valve engines, aluminium pushrods have been used with varying levels of success, but where engine speeds and valve accelerations are high, aluminium pushrods can struggle to be stiff enough. They are used for some sporting applications, but even if they were allowed in applications such as NASCAR Sprint Cup, it is unlikely that anyone would choose to use them.

In engines with pneumatic valve return systems, aluminium is a common material for pistons.

The drive to use more aluminium in valvetrains is important to passenger engine manufacturers, as lower valvetrain mass means a shorter, lower-load valve spring can be used. This minimises valve length (and thus mass), and means it is often possible to lower the height and mass of the engine appreciably. This means that a lighter structure is required to mount the engine, and therefore a lighter crash structure. The car, which is now much lighter, performs better in terms of transient response. The valvetrain and cranktrain components (particularly piston assembly mass) are very potent in this respect, allowing significant weight savings to be made in terms of engine mass as well as overall vehicle weight. 

Written by Wayne Ward

We use a number of different metals as a matter of routine. Pistons are almost always made from aluminium, crankshafts are normally made from steel, and camshafts are fashioned from steel or cast iron.

Superalloys are an odd fit in a race engine. They were developed for high-temperature use in gas turbine engines, since the operating envelope of conventional materials places limits on the performance of turbine engines. Superalloy development therefore continues, as new designs of turbine engines for transporting people and goods around the globe would simply not have been possible using ‘conventional’ or even ‘old’ high-temperature materials.

There are a few genuine applications though for high-temperature materials in any race engine. In supercharged and turbocharged engines, the exhaust valves are probably the only components that need the high-temperature properties that superalloys offer. In such engines, exhaust gas temperatures are very high and the mass flow rate of the burnt gas makes the rates of heat transfer to the valve also very high.

In many naturally aspirated engines, we can often find components made from superalloys in very specific applications – the most common example is undoubtedly that of con rod bolts. These fasteners are extremely highly stressed, and the consequences of their failure are usually catastrophic and very expensive. Almost all con rods for bespoke race engines, and many aftermarket rods for production engines adapted for racing, will be supplied with superalloy bolts or studs. Various superalloy materials are used for con rod bolts – Inconel 718 is quite commonly found, along with others such as MP35N and MP159 and Custom Age 625+, which is similar to Inconel 625.

Similarly, we can find superalloys such as MP35N or MP159 used for other critical fasteners, such as those used to secure main bearing caps to attach the cylinder head to the engine block. These can’t be described as high-temperature fasteners; it’s more a case that the strength of these materials, and their fatigue and corrosion resistance, allow engines to function reliably. In many cases it isn’t possible to use larger fasteners fashioned from less capable materials to cope with the forces and stresses involved. In the case of production based engines in particular, there simply isn't sufficient material surrounding the fastener to allow a larger stud.

Superalloy materials themselves are expensive, and making bolts and studs from them can be costly too, as the material is very hard and therefore difficult to thread roll. It may be necessary to heat the material and to use special thread rolls in order to form the threads properly.

The other main application for superalloys is for valves in naturally aspirated engines. While the engines don’t appear to warrant such expensive materials in terms of temperature resistance, superalloys do offer much higher strength than the austenitic steel or titanium alternatives.

Written by Wayne Ward

The property that makes titanium able to give a slight power increase is thermal conductivity, or rather its unusually low thermal conductivity. At around 6.7 W/mK (46.5 BTU-in/hr-ft²-°F) it is much lower than most of the alloys we would use in race engine construction. 

So, how does this translate to power? Well, the effect is subtle but the answer is simple. If we reduce the temperature of the inlet system, the incoming charge picks up less heat and therefore arrives in the combustion chamber at higher density. Higher density equates to more mass in a given volume, and more trapped mass leads to more output (assuming no loss of combustion efficiency).

One way to do this is to decrease the heat transfer along the inlet tracts. The hottest part is closest to the combustion chamber, and wherever there is a joint between inlet components, we have an opportunity to limit heat transfer. A gasket is a very simple way to achieve this, and many polymers have a thermal conductivity very much lower than titanium. However, the fasteners offer a heat path, and it is worth using titanium fasteners to reduce thermal conductivity across the joints in an inlet system. Where titanium threaded fasteners are prohibited (as in Formula One), titanium washers or bushes combined with any one of the allowed fastener materials will also make a useful difference. Having been involved in a project to reduce inlet temperatures a few years ago, the simulation results were encouraging.

So, how can we reduce aero drag? Well, if we move to the opposite side of the engine, we can use the same strategy to reduce heat transfer into the engine from the hot exhaust system. Having seen the results of attempts to develop a solid ceramic exhaust gasket for a race engine to achieve this effect, I’d recommend using something more conventional. Using a titanium fastener and/or washers to reduce the flow of heat into the heads necessarily reduces the amount of heat the engine needs to reject to the air via the water system. Therefore, if the engine water system is fully optimised, smaller radiators, smaller cooling ducts and less airflow is required.

It is unlikely that swapping a few fasteners is going to allow you to reduce radiator size to any noticeable extent. However, combining lots of small measures to reduce heat rejection from the engine can lead to a small but useful improvement.

Written by Wayne Ward

There are some drawbacks with such materials though. As strength increases, it becomes more difficult to machine the component from the material in its hardened state. There are manufacturing processes that allow the machining of such hard, high-strength steels, but these can present their own pitfalls. Some of the processes that come under this heading include grinding, hard-turning and electro-discharge machining.

The most common way of producing components from very high-strength steels is to machine the component before hardening. However, if the component is finish-machined, the distortion resulting from the harden and temper process can often render the component scrap. If is very often necessary to anticipate an amount of distortion and leave an amount of machining stock on important surfaces to be removed after hardening and tempering. Again we are faced with a requirement to remove hard material, but on a limited scale. Grinding is often used for such processes.

There is though an alternative to machining hard materials and coping with high levels of distortion. Maraging steels are easy to machine and can be hardened without significant distortion. Conventional steels are heated to a critical temperature (called the austenitising temperature) and then rapidly quenched. This is the stage at which distortion is likely to occur. Not only are significant machining stresses which can cause distortion relieved, but the highly uneven temperature distribution, particularly during quenching, can introduce high and uneven levels of residual stress also.

After hardening and tempering, the steel can often only be machined using abrasive processes such as grinding, or requires much slower material removal rates using conventional machining methods such as milling and turning. By removing much of the material from a billet before hardening, the overall manufacturing time is reduced. The machining operations prior to hardening are partly responsible for distortion during the hardening process.

After quenching, the steel is mainly composed of a very hard and brittle phase called martensite. This requires tempering to give us a more ductile and fatigue-resistant steel. However, in doing so, we lose some of the original as-hardened strength.

Maraging steels, following austenitising and quenching, also have a martensitic structure but the martensite is soft and ductile and can easily be machined. They can even be welded relatively easily, which is certainly not the case with conventional hardened steels. After machining, when we want to produce the final level of strength in the material, the steel is ‘aged’. The ageing process is carried out at a relatively modest temperature (in the region of 480 C/900 F), and distortion is consequently lower as the thermally induced stresses are much lower.

The production of a soft phase after initial heating and quenching, followed by the development of the final hardness in the material, is similar to the way heat-treatable aluminium alloys behave.

I did a project to investigate the possibility of simultaneously ageing and surface hardening components made from maraging steels, and there were a number of processes that proved successful in this respect. Maraging steels may not be cheap to buy, but for some components the production process is simplified sufficiently that such materials prove to be economical. They are available with tensile strengths to 2400 MPa (350 ksi).

Written by Wayne Ward

Although the percentage of polymers in race engines and transmissions is low at present, it is likely to increase in future. Also, the increasing use of electrical machines to improve maximum performance and decrease fuel consumption means effective insulators are needed for the battery, power electronics and electric motors, and the cables connecting them all together. We have had electric hybrids in racing for around 15 years now, from the Panoz Q9 in the late 1990s to the current Formula One and Le Mans cars. In order to allow the tight packaging of high-voltage components, a lot of insulating components are needed to prevent these voltages jumping gaps or creeping along non-conductive surfaces to uninsulated sections of conductors.

High-temperature polymers can be used to good effect in components such as inlet trumpets, and where a lot of components need to be produced, it can be economical to replace machined components with lightweight, thin-walled moulded parts from materials such as PEEK (poly-ether-ether-ketone) stiffened by the addition of fillers such as carbon or glass in powder form. Polymers lend themselves not only to machined components but injection moulding as well, so they can be considered as a replacement for machined, cast and moulded components. Even though PEEK is expensive for a polymer (and expensive in general) and the mould tooling is also expensive, the small amount of material waste makes the process of injection moulding attractive. Although ‘virgin’ PEEK is a good candidate for higher temperatures, for those applications requiring better properties there are special grades of PEEK which have improved high-temperature capabilities.

PAI (poly-amide-imide) polymers have been used successfully for engine components and as insulators in racing hybrid systems in the past. It offers improved properties compared to PEEK, but comes at a higher cost.

There are some materials though that put even PEEK and PAI in the shade in terms of temperature capabilities. Among these are PI (poly-imide) polymers, and these have been successfully used inside top-level motorsport engines, although this may only have been for testing. Such materials can be used up to 900 F (482 C). 

Beyond this, PBI (poly-benz-imidazole) is a polymer which can be used with short-term exposure to 1000 F (538 C). ]

While such specialised polymers are very expensive and lack both strength and stiffness compared with metals, their properties – especially as electrical and thermal insulators – mean they will find increasing use in racing power units. The temperature capabilities are good enough to cope with everything we are likely to want them to cope with in the short to medium term.

Fig. 1 - The Panoz Q9 was an early example of a racing electric hybrid; these systems require high-temperature polymers for many electrical insulators

Written by Wayne Ward

In a previous RET-Monitor article giving an overview of the use of composites in the racing powertrain, some of the potential advantages of a composite main case were mentioned. These were not simply concerned with mechanical or physical properties; the complex manufacture of a composite case means there are inherent risks compared to a metallic case. Without wanting to over-simplify the manufacturing route of a metallic case, it is generally a question of making a casting and then machining it. A composite case is much more laborious and requires much greater human input. There is a large number of individual plies that are manually positioned, having been automatically cut, and there are a lot of metallic ‘hard points’ to be incorporated into the composite case which then require machining.

Where metallic cases are essentially homogenous, composites have the advantage that their properties can be ‘tuned’ to achieve a desired result. For example, this can be through judicious use of unidirectional fibres to increase stiffness in a given direction. This is the main mechanical benefit of a well designed composite case, but it is not the only potential improvement over one made from metal. Composites have a low thermal conductivity compared to metals, so the transmission of heat through the walls of the main case is reduced.

Some time ago I discussed the merits of composite transmission cases with an engineer from a now-defunct Formula One team. He said the decrease in thermal energy being transferred through the case resulted in a net reduction in cooling requirement. There is clearly going to be lower heat rejection through the walls of the case, which one might expect to lead to increased transmission oil temperatures and increased cooling requirements. However, this was more than compensated for by a reduction in the amount of heat absorbed by the transmission from nearby hot components, notably parts of the exhaust system from which the transmission case was not sufficiently protected by heat shields. Of course, the amount of heat transfer to the transmission oil through the walls of the case is affected greatly by the proximity of the exhaust, and the current regulation requiring the exhaust to exit on the top deck of the rear bodywork might reduce the amount of heating that would affect the transmission oil.

With the adoption of turbocharged engines in Formula One from 2014 onwards, plus some high-powered electrical machines being used to increase efficiency, we might expect the matter of heat transfer to play a greater part in the selection of transmission case materials in future.

Written by Wayne Ward

One material that engine and transmission design engineers choose in order to achieve this aim is aluminium. There is a bewildering array of aluminium alloys, with many available commercially and with several choices of heat treatment. Naturally, with larger budgets and in aiming for the very best performance from every component, the choice available is less limited by finances. We are often able to choose the strongest, most reliable and least dense or stiffest alloys. As with so many things, motorsport feeds after the aerospace and defence industries have eaten; we have access to materials that have been developed by them. It should come as little surprise then that many of the aluminium stockholders supplying motorsport also have a lot of aerospace and defence business.

Still widely favoured for race engine pistons, among other things, 2618 is an alloy originally used many decades ago for aero engine pistons. Other traditional 2000-series alloys enjoy widespread use in motorsport, as they maintain a useful percentage of room temperature strength to temperatures in excess of 150 C. Examples of widely used 2000-series alloys are 2014 and 2024; the former is widely used in Europe while the latter is more popular in the US.

A relatively new class of aluminium alloys are those containing lithium. Enjoying high strength and stiffness, combined with low density, these are a premium material, but one that can offer significant advantages in certain applications. They can offer up to 10% increase in elastic modulus, and up to 10% decrease in density, compared to conventional aluminium alloys. In terms of specific modulus – that is, elastic modulus divided by density – they offer a huge advantage compared to other aluminium alloys. Most such commercially available alloys are classified in the 2000 series – 2050, 2090 and 2099 are examples. Alloy 8090 is the main commercial Al-Li which is not in the 2000 series; it contains magnesium and is among the least dense aluminium alloys.

The high-strength 7000 series alloys remain popular for applications where temperature is not an overriding concern. Adequate up to moderate temperatures, such alloys have very high strength. 7075 is widely used for high-strength components, as it has an ultimate tensile strength of >500 MPa (72.5 ksi) in a number of commercially available tempers. 7068 is the strongest of the commercially available aluminium alloys, with a minimum tensile strength approaching 700 MPa (101 ksi). It is not yet thought however to have found widespread use for engine and transmission components.

Moving briefly from wrought aluminium alloys to cast products, aluminium is used widely for major structural components in race engines. The typical A35x alloys which have been used widely in motorsport and aerospace continue to be popular. There are though new, very high strength casting alloys under development for racing applications, having been used initially in aerospace. The A20x family will find increasing use in race engines.

Written by Wayne Ward

However, for the purposes of this article, I want to discuss the powder metallurgy method of producing very high quality wrought-steel alloys. Such alloys are now becoming much more widespread, and can offer significant advantages over conventionally manufactured alloys, albeit at a cost penalty. The production method was discussed in the recent article on metals in Race Engine Technology magazine, and in a RET-Monitor article in 2012. We will look here at the production process in more detail, hopefully giving the reader a better understanding of why the steels might offer an improvement, and why the materials are so much more expensive.

As mentioned in the previous RET-Monitor article, powder metallurgy steels offer improved fatigue and fracture properties for the same composition compared to conventionally manufactured steels. Another stated benefit is that they allow more highly alloyed materials to be produced than with conventional production methods.

The material from which the powder is produced is first melted under vacuum. This is done to prevent oxidation of the surface and the danger of oxide contamination spoiling the product. There may be an ‘extended process’ where any impurities are removed through solidification and re-melting; many of the conventionally manufactured steels of very high quality are re-melted products. Where we see the acronyms VIM, VAR, VIM-VAR and ESR applied to steels, these are re-melted products. Such techniques produce clean steels that exhibit better durability compared steels that are not re-melted and of the same nominal composition. In previous ‘Focus’ articles on specific components in Race Engine Technology magazine, suppliers of very highly stressed steel components will often specify single or double re-melted steels because of their improved fatigue strength.

Once the material from which the powder is to be produced is selected, atomisation takes place by melting the material and passing it into high-velocity jets of nitrogen, which break up the stream of molten metal into very small spherical particles that solidify very quickly. The advantage of such rapid solidification is that each particle thus produced is very uniform, with very fine precipitated carbides. Where cooling is slower, the carbides can tend to segregate into larger carbides rather than being finely and evenly dispersed.

The collected powder is put into steel ‘cans’ which are then evacuated and sealed, commonly by welding. The sealed containers are then placed into a furnace where hot isostatic processing (HIP) takes place. You may have heard of the HIP process in relation to castings, where the process results in better quality and stronger castings owing to a lower level of porosity. In the powder metallurgy process, HIP compacts, joins and densifies the powder into a solid, fully dense ‘compact’.

Once the HIP process is complete, the can is removed and the semi-finished material is processed by conventional means such as forging, rolling and drawing. The cost of the process means that most steels produced by this method are high-value tool steels. The technique is also used for superalloys.

Written by Wayne Ward

There are of course other applications for this technology and the increasing range of metallic and non-metallic materials now commercially available. Recently, on the eve of the Autosport Engineering trade show in the UK, I was shown a hollow poppet valve with a very detailed internal structure. Hollow-headed valves are prohibited in many race series such as Formula One. The valve I was shown was produced by metal laser sintering in a high-temperature material. Its internal structure would be impossible to produce by any other production method.

Increasingly, rapid prototyping methods have shown themselves to be a serious production method for niche applications and one-off requirements. Surely the day will come soon when small non-racing automotive producers will look to such methods for producing components in passenger cars, motorcycles or freight vehicles. Much the same happened with titanium valves and con rods, which were once the preserve of racing teams; both are now used in passenger cars and motorcycle engines.

The valve I saw might prove to be very suitable for the new 2014 Formula One rules, or perhaps for the new Le Mans engines also set to début in 2014, but it was actually destined to be tested on a commercial diesel engine. Again, another regulation introduced initially to save cost will only sideline motor racing when it comes to certain areas of component development, putting high-level motorsport one step behind basic automotive research projects.

The prospect of relatively inexpensive hollow-headed valves is something that a lot of engine design and development engineers will find enticing, and I am sure many of them will look towards this manufacturing technique once the materials and methods are proven to produce parts which not only have the desired static mechanical properties but which can also demonstrate adequate fatigue life.

In terms of materials, there are developments going on all the time. Metal matrix composite powders are being investigated with a view to ‘tuning’ the performance of some relatively ordinary powders that are already commonly available.

When it comes to introducing these laser-sintered materials and their eventual success, the limiting factor will be the creativity of the design engineers. Those who are too heavily encumbered with a lifetime of ‘design for manufacture’ will perhaps find it difficult to let go of these constraints and design what they actually want. Often the comment from producers of rapid prototype parts is that the major failing of designers when designing for the new manufacturing technique is that they simply design something very close to what could be produced by conventional methods – for example, a coolant passage in a rapid prototype part designed as a series of straight holes rather than the continuously curved channel, which is better from a flow point of view and which might have been the obvious choice to a complete beginner with no preconceived notions of how a part should be designed.

Written by Wayne Ward

There are some very obvious applications of superalloys in motorsport, especially where we see high temperatures. The trend in passenger vehicle engine design towards smaller forced-induction engines is being mirrored to some extent in racing: forced induction leads to higher combustion and exhaust gas temperatures.

If we follow the gas path through a typical engine, the exhaust valve is the first component which is routinely made from superalloy materials. As the valve is opened, the ratio of surface area to volume is high, as is the speed of the gas past the valve. These two facts give rise to high heat transfer coefficients from the exhaust gas to the valve, and its rapid heating. While naturally aspirated engines can generally cope using titanium or steel valves, forced-induction engines very often require materials that are better able to withstand the high temperatures. Both Inconel and Nimonic alloys are options for valve materials which are commercially offered by racing valve manufacturers.

Exhaust systems made from superalloy materials are also commonly offered by specialist manufacturers. Here, the strength at temperature is an important consideration, and can mean that superalloy materials offer the lightest possible reliable exhaust system. These are routinely used in many racing classes from Formula One, through NASCAR Sprint Cup to motorcycle racing.

In exhaust system design, considerations of creep are also important. Creep can be summarised as either the increasing strain over time under a fixed stress, or alternatively the reducing level of stress required to maintain a fixed strain in a component. Superalloys are very creep-resistant and are specifically developed with this criterion in mind. Materials which prove themselves over the course of a short dyno test may fall short of expectations when running at the circuit, with exhaust systems ’sagging’ after a while, potentially coming into contact with parts from which they were initially separated by a generous air gap.

For turbocharged engines, where strenuous efforts are made to retain heat in the exhaust gas flow, the turbine wheel has to cope with very high temperatures. This is the application with the greatest similarity to the aero engine applications for which such materials are usually developed, although turbochargers are radial flow devices rather than axial flow. Again these components are routinely made from superalloys. Although ceramic and titanium turbine wheels offer advantages in terms of low inertia and therefore improved transient response, they have proven to be unreliable in the past.

Superalloys are also very corrosion resistant, so much so that there is little difference in the fatigue properties between their use in air and sea water for some alloys. This resistance to corrosion may appear not to be applicable to most race engine components, but there is certainly opportunity for corrosion from the fluids in a race engine. Highly stressed, high-strength fasteners can be susceptible to environmental hydrogen embrittlement over long periods of time. Highly stressed fasteners are often made of superalloys, with con rod bolts being available in a number of different materials of this type.

Written by Wayne Ward

However, it appears that people are still looking at the possibilities for titanium crankshafts, despite the relative drawbacks of using the material. Moreover, components have been made in a titanium metal matrix composite (MMC).

Both titanium and any form of MMC are banned under the FIA's Formula One regulations for crankshaft manufacture, and crankshafts made from them are certainly not destined for Formula One or any other kind of top-level motorsport. These parts have run with limited success in a real engine, and this has to be considered a step forward. The project may be a dead end, with no real push for further development, but it may also be a significant first step towards bringing the technology into everyday use.

The research has been run by a large multinational roadcar manufacturer, so we have to at least consider the fact that such materials are being developed for use in passenger cars. The advantage of using a titanium MMC material is the increase in modulus owing to the reinforcement, meaning that bearing diameters are much closer to those of a steel crankshaft than would be possible using a conventional, non-MMC titanium material. The frictional losses are therefore lower with the stiffer material, owing to the smaller bearings.

The material manufacturer has told me that the crankshafts ran for some hours in a test engine, and that it eventually failed owing to a surface-related mechanism. This should come as no surprise to people who have worked with titanium, a material whose propensity to fail due to surface damage in even lightly loaded sliding contacts is well known.

To reduce the likelihood of failure, the crankshafts were DLC-coated. However, such coatings, while hard and having low friction coefficients, are only very thin, and to be successful they need to be properly supported. Having a relatively hard and brittle thin coating placed on top of a comparatively low modulus material that is prone to surface damage often leads to premature coating failure. Besides DLC, there are a number of coatings that can be applied to titanium, although none are generally applied in layers thick enough to withstand highly loaded sliding wear, especially when lubrication is marginal, as is the case at engine start-up.

Written by Wayne Ward

Aluminium combines low mass and reasonable strength, and it finds use for pistons and all sorts of other components. The specific modulus or specific stiffness of aluminium is typical for metals. If we divide the elastic modulus of a metal by its density, for most common metals that we use in a race engine we will normally arrive at a similar number. Steel, titanium, aluminium and magnesium all generally fall within the range 24-28 GPa/(g/cm3) range.

If we want to use something with a greater specific modulus, we generally have to look at something fairly exotic, such as a metal-matrix composite (MMC), where particulate ceramic reinforcement is introduced in order to enhance certain properties.

However, there are a number of commercially available aluminium alloys that differ substantially from the 'usual' properties we would expect of aluminium. Silicon is widely used as an alloying element, and can commonly be found in a number of wrought and cast aluminium materials used in race engines. Piston materials with around 12% aluminium are commonly used for bespoke forged racing pistons, and higher percentages of silicon are used for some cast racing pistons.

With increasing silicon percentage comes decreased ductility, and materials with greater than 20% silicon are felt to lack the elongation required for a piston, hence we find that cast high-silicon pistons are rarely more than 19% silicon. However, one advantage of the higher silicon alloys is their lower thermal expansion coefficient. This allows the engine designer to run tighter cold clearances between the piston and cylinder bore.

Use is made of the relationship between silicon and decreasing thermal expansion coefficient in the manufacture of controlled thermal expansion alloys, and with very high proportions of silicon, these can match steel and even titanium in terms of their low thermal expansion coefficient. There are many applications outside of racing where this very low thermal expansion in a lightweight alloy is of interest, but it is possibly just of academic curiosity to us.

However, these alloys possess other properties that might be of real interest to us. They are much stiffer than conventional aluminium alloys and are also less dense. In fact, this combination of properties would make some aluminium-silicon alloys fall foul of the FIA Formula One limit of 40 GPa/(g/cm3) limit for metallic engine materials, even though they are not an MMC.

In racing applications, the limiting factor for the use of such materials is likely to be lack of ductility. While we generally don't design engine or transmission components to by cycled in the plastic region of the stress-strain curve, there is often some very local plastic deformation, especially on component edges and in areas of contact against stiff adjacent components. It is from such areas that fatigue cracks can initiate.

Written by Wayne Ward

It is true that many modern engines will have some visible carbon fibre-reinforced polymer matrix composite parts, but often such materials are used only for airboxes/plenums and covers. There are a few race engine components that are routinely made from polymers. The most obvious example of polymer components are seals. From PTFE lip seals to elastomer O-rings, there is no viable alternative material. For O-rings and rubber seals, nitrile rubber is often OK, but where rubber seals are exposed to high temperatures or fuels, fluorocarbon rubbers are generally used.

Where stresses are low, polymer parts can be substituted for aluminium parts quite easily in order to achieve weight loss. Both filled and unfilled polymers are suitable for this purpose, and it is common to see these materials used for inlet system components where there is a constant supply of cool air. Cost savings might also be made where component quantities allow the use of polymer processing methods such as injection moulding or vacuum casting.

For small development batches, some 'rapid prototype' methods can produce cost-effective parts for engine use, although such methods are constrained in their use owing to the limited range of materials available. For moving components, there is very little use of polymers. Cages in high-specification ball bearings are commonly machined from polymer materials such as PEEK. Some oil pump suppliers offer polymer pump elements for scavenge pumps. Other minor components such as pressure-relief-valve pistons are good candidates for being made from polymers.

For anything more adventurous than this, we need look no further than the Polimotor engine from the 1980s, which is often mentioned here in RET-Monitor and in Race Engine Technology. RET issue 56 carried an article on the method of producing polymer cylinder blocks, and the SAE paper by Gaudette details some of the components used in this pioneering engine.

The simple 'rule' for this engine project was that components that ran at less than 500 F (260 C) would be considered for manufacture from a polymer or polymer matrix composite material. Most of the polymer components in this engine were filled polymers, and so qualify as composites. While 'composite castings' are being used in small numbers for race engine cylinder blocks again, the moving components in the Polimotor engine are of particular interest.

A number of parts were in some of the most demanding applications. The piston was partly made of polymer material, as was the piston pin. In the valvetrain the use of polymer parts was widespread, with cam followers, valve spring retainers and intake valve stems being tested and raced. The engine was an overhead cam unit, but trial parts were made for overhead valve (pushrod) engines too, with rockers and lifters being produced.

It is unlikely that we will see such heights of engineering reached again, as we have come to rely very heavily on the strength, stiffness and temperature resistance of metals for moving engine components. However, the work done almost 30 years ago perhaps points to the potential for polymer components to be used for a wider variety of applications.

References
Stowe, J., Race Engine Technology 'Insight' article, issue 56, August 2011
Gaudette, E.P., "Plastics within the Internal Combustion Engine", SAE Paper 850815, 1985

Written by Wayne Ward

However, this has been slowly changing for some time. Currently there are few applications of composites in race engines, although development is limited in many cases by regulation. In the past decade we have seen various engines sporting carbon-reinforced polymer cam covers, water pipes, heat shields and various other non-structural covers.

Polymer matrix composite (PMC) materials - where, as the name suggests, a reinforcement is used in conjunction with a polymer matrix - are becoming more popular all the time. The mid-1980s Polimotor was a fantastic piece of design, incorporating all manner of stressed engine components in reinforced polymers. From 'composite castings' through piston skirts, con rods and valvetrain components, this competition engine didn't receive the ongoing development budget it really deserved.

Almost 30 years later, it still stands as the pinnacle of polymer composite materials development for engines*. The technology of composite castings is still being actively developed, and recently large structural components - including cylinder blocks for racing - have been produced in a carbon-reinforced composite material, and these are available commercially.

In terms of polymer composite transmission cases, the story is very different. Formula One teams have used full carbon fibre gearbox main cases for over a decade now, and various parts of Formula One transmission cases have been made from the material for more than 15 years. As with the fully PMC chassis, the technique of using carbon fibre for transmission cases was pioneered by John Barnard. Strangely, the technique seems to ebb and flow in terms of popularity; various teams have produced and used carbon fibre gearbox cases and then reverted to metallic cases again.

The case for metallic cases is still strong, as many people understand metallic materials and the processes required to make a sound structural casting. The process of making carbon fibre reinforced gearbox cases is difficult and requires great attention to detail in the design and manufacturing stages to ensure success. However, the rewards are potentially great. If properly designed, a stiffer structure is possible for the same mass as a metallic case, and high stiffness is one of the primary aims of the design of a modern single-seater racecar.

There is a widespread misconception that, where a part can be manufactured successfully from PMCs, it will perform better than a metallic part. This is not necessarily true; because of their structure the load transfer through composites is not straightforward, and an understanding of how these parts behave in practice is the key to designing, manufacturing and using structural composite components.

As the saying goes, you need to choose your battles wisely - that is, don't try to replace everything metallic with PMCs; choose the parts that are within your capabilities and which will bring most benefit.

* Stowe, J., 'Focus' article on the Polimotor, Race Engine Technology, issue 56, August 2011

Fig. 1 - The Polimotor project of the early 1980s remains the pinnacle of polymer-matrix composite use in engines. The yellow components were made of polymer-matrix composites

Written by Wayne Ward

Copper alloys also find extensive use in some types of 'shell' plain journal bearings, as used for crankshaft main and crankpin bearings and some camshaft bearings. However, in this case, the copper alloy is not intended to come into contact with the rotating component; it is simply an intermediate layer, normally around 0.30-0.38 mm (0.012-0.015 in) thick between the backing and the soft metal alloy (often called white metal or 'Babbitt') that forms the actual bearing surface. This type of bearing construction is known as 'tri-metal' and has been described in previous RET-Monitor articles by John Coxon and Eric Smart.

If we have a strong steel backing and a suitable bearing surface, why do we need an intermediate layer of copper alloy? There are several reasons. The bearing surface itself is made of an overlay material with very low shear strength and high wear rate. It has to be made thin in order to be able to support the forces imposed on it. The copper alloy layer underneath has to be strong enough and stiff enough to support the loads, while being compliant enough to allow sufficient movement and flexibility for the crankshaft. If we were to plate the overlay directly onto a steel backing, any flexibility in the crankshaft would quickly lead to high edge loading and an increase in friction and wear. Another reason for the use of a copper alloy is the slight 'embeddability' it gives the bearing, if the overlay material should fail, by being able to tolerate and capture particles in the oil.

The copper alloy is alloyed with lead and, in some cases, tin. The primary purpose of the lead is to provide a level of lubrication, and the tin is there to provide strength. The lead in the copper alloy also gives some degree of embeddability. Lead-free bearings are widely used in the production car world, and they have been widely discussed in RET-Monitor by Dieter van der Put and Ian Bamsey.

However, a discussion with a leading motorsport bearings supplier revealed that it supplies no bespoke lead-free bearings for race engines. The inherently greater flexibility of race engine crankshafts and camshafts, and their high duty cycle, means the lubricating properties of the intermediate layer are more often called upon. While the lubricity of the copper-lead alloy rose with increasing lead content, the strength falls; lack of fatigue strength of the copper-lead alloy may become a problem when the soft bearing metal has worn away or has fatigued. There is a whole family of copper-lead-tin alloys with differing strength and friction characteristics.

There is a degree of 'comfort' and familiarity with the use of copper-lead alloys in bearings for race engines. There may come a time where lead becomes outlawed for bearings, in which case copper-bismuth alloys are one likely replacement.

Fig. 1 - Sometimes we see more of the copper alloys in tri-metal bearings than we would like

Written by Wayne Ward

Despite the fact that steel-making is a very mature technology, steel materials still enjoy a lot of r&d. One technology leading to better performing steels is that of powder metallurgy. There are a number of reasons why it is being used in steel production. One is that it allows materials to be formulated which cannot easily be made using more conventional ingot metallurgy methods. However, a number of powder metallurgy tool steels are made to existing standards and compositions. So, why would a steel company go to the trouble and expense of melting a good piece of tool steel, only to reform it as a powder metallurgy material?

Producers of powder metallurgy steels say the microstructure of the steel is far more consistent, and that the steels produced by this method are more ductile and less prone to distortion during heat treatment, even compared to steels of the same composition produced conventionally.

The method by which powder metallurgy steels are made follows five steps:

" Production of the powder from a melt
" Mixing and blending of powders, where applicable
" 'Canning' of the powder, by filling a steel box with powder and welding it shut
" Consolidation under high pressure and temperature in a hot-isostatic pressure process, and
" Processing of the billet thus produced

The production of the powder is a critical step, and is the subject of much development itself. The production of ever-finer powder stock has its own benefits - it stands to reason that the size of the powdered material limits the size of any non-metallic inclusions or defects. Defect size affects the fracture and fatigue behaviour of materials: those produced from finer powder have a fundamentally better chance of being more durable than those made from a coarser one.

The concept of producing very highly alloyed materials by powder metallurgy methods is attractive. The trends in properties predicted by adding certain elements is limited by the effect of carbide coarsening and segregation in conventional ingot metallurgy materials.

In a study of the high-cycle fatigue behaviour of conventional and powder-metallurgy steels*, fatigue-crack initiation for ingot steels was found to begin at large carbide particles or clusters of carbides, whereas for powder metallurgy grades, fatigue cracks were initiated at non-metallic inclusions. The report notes that the powder metallurgy materials have significantly higher fatigue strengths than their ingot metallurgy counterparts. Powder metallurgy methods mean that the dispersion of carbides within the material is very even, avoiding carbide segregation and coarsening.

These materials would seem to offer the possibility for lower-mass components in motor racing, where equivalent powder metallurgy grades for existing materials are available, and the chance to produce materials that are not possible by ingot metallurgy methods.

* Danninger,H., Sohar, C., Gierl, C., Betzwar-Kotas, A., Weiss, B., "Gigacycle Fatigue Response of PM versus Ingot Metallurgy Tool Steels", Materials Science Forum, vol 672

Written by Wayne Ward

In the past few years, there have been a number of companies who have been able to offer some very 'useful' metallic materials with strengths to rival those of commercially available cast and even wrought alloys. In general, with only a few notable exceptions, the metallic parts of race engines are made from aluminium, titanium and steel. Owing to the pressure to produce ever-higher performance from an ever-decreasing engine mass, parts naturally become more stressed. Readers with an interest in this article would do well to read John Stowe's excellent article on the subject in issue 59 of Race Engine Technology.

In the initial stages of adopting a new manufacturing technology, the first use that many people will find is as a straight substitution for an existing material and manufacturing method. In general, they will produce a design very similar to that which they might have produced with conventional methods. For the 'rapid prototyped' component to be competitive on weight, it will need to have comparable strength to the part it replaces. Later, as the real benefits of the manufacturing method are taken account of, and designers feel less constrained by their experience of what can be produced conventionally, considerable mass savings are likely to be found.

However, before people are happy to use rapid prototyped metallic components, they will check the material strength and, if available, fatigue properties. Fortunately, there are now a number of materials that offer a real prospect of making top-level race engine components, even if their costs are very high at present.

There are two commercially available aluminium materials, based on aluminium-silicon alloys. Containing 10% and 12% respectively of silicon, these are similar to some high-strength casting alloys, and compare well on strength. Aluminium is of great interest, as many 'static' components are produced from this material.

Titanium Ti-6Al-4V shows great promise as a rapid prototyping material, with strength figures to rival wrought bar, but the manufacturing method has the ability to produce parts of fantastic complexity. Other titanium materials are also commercially available in suitable powder form ready for manufacture.

With two Inconel alloys and two cobalt-chromium materials, the high-strength 'superalloy' base is covered, offering the opportunity to produce hot components for exhaust system or possibly turbocharger 'hot-side' use.

High-strength steels are also very well represented; precipitation-hardening steels such as 15-5 and 17-4 offer a 'medium to high strength' option. Higher strength steels are also covered, with both hot-work tool steels such as H13 and maraging steel being commercially available.

There are three barriers to such methods becoming widely used for race engine components though, other than for very specialised parts - the cost of the components, the size of component that can be made, and sometimes the speed of the 'build'.

Fig. 1 - Exhaust production using additive manufacturing methods has been used by a number of companies

Written by Wayne Ward

Such high-temperature materials are not just fancy steels; in comparison even to very highly alloyed stainless steels they contain very little iron, if any at all. Internal combustion engine development has benefited from the use of such materials in a number of applications.

Inconel is a trade name that encompasses a wide range of high-temperature alloys whose major constituents are nickel and chromium. They are expensive, not only because they are very 'clean' and well controlled in terms of chemistry, but because they have a high proportion of 'strategic elements' - those that are deemed to be in short supply and of great economic and political importance. The UK government has identified a number of elements that are strategically important*, and the list includes both nickel and chromium, both of which it deems to have 'limited availability' and a potential 'risk to supply'.

The most commonly discussed use of Inconel in connection with race engines concerns its widespread and growing use in the manufacture of exhaust systems. As teams come to realise that Inconel exhaust systems can be both light and durable, they are becoming more popular, with the use of the material 'trickling down' from the highest budget series, where minimum mass is the goal, to lower-budget series, where durability is important. The most commonly used Inconel alloy for exhaust system manufacture is alloy 625.

However, there are a number of other applications for Inconel materials. They are quite widely used for highly stressed fasteners and find use as con rod bolts, particularly in alloy 718. While this application does not take advantage of Inconel's high-temperature capabilities, the tensile and fatigue properties at room temperature are impressive, and the material is much more resistant to corrosion than steel materials of equivalent strength (230 ksi/1600 MPa).

There is one use of Inconel alloys in the engine, however, that does take full advantage of the high-temperature fatigue strength of the material. Inconel 751 is used for valves where high temperatures are experienced; this alloy is especially popular for supercharged or turbocharged engines, where exhaust valve temperatures are much higher than would be the case for a naturally aspirated engine.

In a naturally aspirated race engine, the use of austenitic steels such as 21-4 is common, as is the use of titanium. However, under the harsher conditions imposed by running an engine with significantly increased mass flows and temperatures, these materials cannot operate for very long. While alloys such as Inconel 751 are denser than either austenitic steels or titanium, their use for valves is becoming increasingly necessary.

* House of Commons Science and Technology Committee Report, "Strategically important metals" (http://www.publications.parliament.uk/pa/cm201012/cmselect/cmsctech/726/726.pdf)

Fig. 1 - Inconel valves find use in supercharged and turbocharged race engines

Written by Wayne Ward

Besides titanium's obviously attractive property of low density, its elastic modulus is the other property that makes it a good candidate material for fasteners, both of the internally and externally threaded varieties. The use of nuts with lower modulus than the male fastener is known to reduce the stress concentration effect at the first thread, and improves the distribution of load over the length of the engaged threads. Where high-modulus materials are specified for both internal and external threads, one way to achieve the same effect is to use combinations of male and female parts with very slightly differing thread pitches.

When considering the design of a stud or bolt used for cyclically loaded fastener, it is important to consider both the fastener stiffness and the stiffness of the parts being clamped. A simple formula involving these quantities dictates how the service load is shared between the unloading of the joint, and the extra load borne by the fastener. This relationship has been covered in one of the early RET Monitor articles on fasteners and in a past article in Race Engine Technology magazine*.

The smaller the stiffness of the fastener is compared to the stiffness of the joint, the less of the service load that is borne by the fastener. Ideally, what we want from a fastener material is for it to be strong - fatigue strength is the significant strength in a cyclically loaded fastener - and to have low stiffness. Titanium can score well here, and its lightness is a bonus, although that shouldn't come as a surprise. Most metallic materials fall within a pretty narrow range of specific modulus (modulus divided by density) and so any material with a low modulus is likely to have a low density. There are some notable exceptions to this 'rule of thumb', such as beryllium, but most common aluminium, magnesium, titanium and steel alloys we are likely to commonly use have very similar specific modulus values.

There are some technical problems though with the use of titanium as a male fastener. Its tendency to gall at low levels of load when sliding means it needs to be installed with special grease, or needs to have its surface treated to prevent the problem, especially where it is used in conjunction with a titanium nut. However, the problem is far from insurmountable, and racing motorcycles of 20 years ago were festooned with such fasteners throughout the engine and chassis, as are many racecars, motorcycles and boats today. It seems strange that they are now outlawed in bespoke race engines at the highest levels of motorsport, but are affordable to low-budget racers.

*Race Engine Technology, issue 41, September/October 2009

Fig. 1 - Was motorcycling in the early 1990s a richer sport than Formula One is today? Titanium fasteners are banned from Formula One engines, but were affordable for endurance racers such as this beautiful Honda RVF750

Written by Wayne Ward

While cast aluminium is likely to maintain its dominance for structural castings, there is widespread use of wrought aluminium materials for all manner of highly stressed race engine components, from the almost ubiquitous aluminium piston, to the less common aluminium con rods, through to billet cylinder heads, blocks and sumps. There are many other uses for wrought aluminium, from valvetrain components to pump housings and internals.

There are ever stronger and lighter alloys being brought to our attention in motorsport, often trickling down from the aerospace industry. These are available in a number of different strength conditions, as specified by the H or T designations. Materials that don't respond to heat treatment, such as the 5000 series alloys, are given different levels of strength by the use of deformation processes. Those that do respond to heat treatments, such as the 2000, 6000 and 7000 alloys, are classified according to both heat treatment and other deformation treatments. A cursory glance at an aluminium supplier's webpage revealed that they can supply 7075 alloy for example in 17 different grades.

Even within a given strength grade, we may not get what we expect. A supplier of complex aluminium components (specifically for motorsport), recently explained that he makes some components from blanks that he manufactures in-house from a laminated structure made from rolled plate, effectively producing a bar from a stack of round plates. These are bonded together to form bars of the correct length, which are subsequently machined.

There were a number of reasons for this, he said, but one of the most important was that the material in the centre of round-drawn bar does not undergo the same amount of deformation as the material at the outside. Consequently, the material at the centre of the bar is not at the same level of strength as that at the outside diameter of the bar. He explained that when he needs to incorporate highly loaded splines in the centre of the component, only rolled plate consistently has the required level of strength.

Fortunately for him, there is no requirement for the part to take any tensile stress in the axial direction which would load the multiple bonded interfaces. Given the complex manufacturing route, he has turned this to his advantage in other areas to produce a very successful racing product.

As with the example of the directional strength of steels given in a previous article, aluminium products also require some care in the selection of the correct material, taking into account the direction of any applied loads. Again, the processing of the material can give significant differences in strength in different directions, and the maximum loads should be, where possible, aligned with the direction of mechanical deformation.

Fig. 1 - When specifying high-strength aluminium, be sure to align the high stresses with directions having high strength

Written by Wayne Ward

There are certainly some special alloys that can improve on this, but very little that would come close to breaking the FIA-imposed limit of 40 GPa/(g/cc) for metallic materials in Formula One.

If there were no such 'glass ceiling' imposed by regulation, and we wished to maximise specific stiffness, we would probably need to turn to metal-matrix composites (MMCs). As the name suggests, these are materials where a reinforcement material is held within a metallic matrix. Most such materials are based on finely divided discontinuous reinforcement using a number of ceramic materials, although ceramics are not exclusively used as the reinforcement or filler.

A number of these materials are commercially available, with at least one commercially available aluminium MMC exceeding 48 GPa/(g/cc). Where the mass of a component is determined by considerations of stiffness, such materials are capable of realising much lower mass components.

The strength levels of materials can be significantly raised, too. For example, there is an aluminium MMC with an ultimate tensile strength of 1300 MPa (188 ksi) - more than double that of an exotic 'conventional' aluminium alloy, and three or four times that of a typical material. This level of strength is exceptional, even for an aluminium MMC.

However, we should not assume that improvements in specific stiffness and strength are the only gains to be had. MMCs can also offer improvements in other properties - they are used to improve thermal conductivity, even for materials where the conductivity of the matrix is already good, for example in copper or aluminium. While it is common for people to think of ceramics as good thermal insulators, this is a generalisation that does not hold true; for example, the thermal conductivity of aluminium can be improved by adding certain ceramics.

So, where can we find MMC materials used in current race engines? Aluminium MMCs have found use in pistons, con rods, and valve spring retainers. An MMC aluminium connecting rod has been covered in these Monitor articles in the past, with the rod further distinguishing itself by running with neither a bearing shell nor a roller bearing. MMC steel has been used in piston pins, and MMC titanium is thought to have been used in at least one Formula One engine for the same purpose in the past. Prototype MMC inlet valves were tested some years ago, but were not successful.

If Formula One and other big-budget race series such as NASCAR had less restrictive material regulations, we would see more use of MMCs, and their use would possibly become more widespread in other series as a consequence.

Fig. 1 - MMC pistons would perhaps be more widely used if motorsport material regulations were more flexible (Courtesy of AMC-MMC)

Written by Wayne Ward

The use of tungsten as engine ballast where mass and centre-of-gravity regulations exist isn't out of the question, with some manufacturers again building engines very slightly underweight and ballasting them to meet the minimum weight limit. In Formula One, where the minimum height of the engine's centre of gravity is specified, the motivation for ballasting engines is minimal. In the past, it wasn't unknown to have sumps made from materials not normally associated with race engines in order to make the centre of gravity of the engine low.

However, the most common use of tungsten in the race engine is for crankshaft counterweighting, as has been discussed in a number of previous RET-Monitor articles dealing with methods of attachment. The merits of each method of attaching the tungsten to the crankshaft have been discussed, and it has been stated that the most effective method is to replace the conventional counterweight with tungsten, which is bolted to the crankshaft.

In order to prevent the crankshaft and the counterweight parting company (which is not unknown), the bolts need to be tightened to very high levels of pre-load. The bolts are often specially designed and made from special materials. The loads are such that specially processed tungsten material is required in order to prevent yielding of the tungsten under the bolt head, either during initial tightening or in service. Any yielding taking place is likely to lower the level of pre-load in the bolt and can introduce the risk of the bolt becoming loose or of joint separation. Joint separation leads to extremely high service loads in the bolt and increases the chances of the bolt failing through fatigue.

The type of tungsten alloy which is used for the highest strength of counterweight mass is processed by cold working, and commonly by swaging - a form of radial forging. This markedly increases the material's yield and tensile strength, but means that such material is generally most widely available in the form of round bars, and hence requires more machining than the lower-strength tungsten alloys which are available in the form of plates and rectangular bar. There are only a limited number of companies that can supply this material, and the most common outlet for this type of very high-strength tungsten is for military use.

There are a wide range of tungsten alloys available, and the most common alloying element is nickel, which is present in almost all the alloys that you find, with other additions of copper, iron and molybdenum. The various alloys available have a range of densities from 17g/cc to 18.5g/cc (steel is around 7.85g/cc), and it might surprise you to find that it isn't necessarily the densest alloys that are used for high-strength counterweights.

Fig. 1 - A race crankshaft with bolt-on tungsten counterweights (Courtesy of MG Sanders)

Written by Wayne Ward

Beryllium is a very lightweight element, and has some attractive mechanical and physical properties. However, there are health considerations regarding its use. Copper-beryllium alloys are not felt to be a serious health concern, although some companies are looking to limit its use as a precautionary measure. Copper-beryllium alloys certainly don't represent the same health risks as aluminium-beryllium materials, where pure and finely divided beryllium is held in an aluminium matrix.

Aluminium-beryllium is banned in Formula One (which is probably one of the only race series where budgets would justify its use), but copper-beryllium alloys are within the rules. Aluminium-beryllium materials contain a large proportion of beryllium, where the typical copper-beryllium alloys found in use in race engines contain 2% or less of beryllium.

The drive to limit the use of beryllium has lead to beryllium-free alloys being developed, and these have found a number of uses in motorsport, being used as a direct replacement for copper-beryllium alloys in many applications.

A number of copper-nickel-tin 'spinodal' alloys have found use in bearing applications such as valve guides, lifter bore sleeves and so on. These alloys are said to exhibit low wear and friction, and combine low friction with a high level of yield strength. However, they do not possess the high levels of thermal conductivity required for some of the more demanding thermal applications in race engines. Where valve guides are required to remove a large proportion of heat from a valve, we would still expect to find high thermal conductivity alloys in use. The same applies to valve seats which, in most engines, will be responsible for the conduction of most of the heat away from the valve head.

Some copper moulding alloys have found widespread use in race engines where the combination of strength, wear resistance and thermal conductivity is required. The main alloying elements in such alloys are nickel, silicon and chromium. They find current use in race engines as valve seats and guides, valve-lifter sleeves (for pushrod valvetrains), con rod bushes and as bushes in engine where people are looking to replace rolling element bearings, but where no suitable bearing shell exists.

Copper-chromium alloys offer reasonable strength and very high thermal conductivity, and might find use in applications where dissipating heat is an important requirement, such as valve seats, valve guides (especially where sodium-cooled valves are used) and in highly loaded con rod bushings.

In terms of high-conductivity copper alloys being used for valve seats, it is common practice to use different materials on inlet and exhaust valve seats, with a higher conductivity alloy generally being specified for the exhaust valve seat. The two most common grades of beryllium copper in use for these components can both be replaced with beryllium-free alternatives offering very similar or higher thermal conductivity.

Fig. 1 - Valve seats can be replaced with beryllium-free alternatives that offer equivalent thermal conductivity (Courtesy of Performance Alloys and Services)

Fig. 2 - This is the bore of a valve lifter sleeve. Wear resistance is an important property for these parts (Courtesy of Performance Alloys and Services)

Written by Wayne Ward

By the use of directional processing, such as rolling or extrusion, any defects in the steel are elongated. Inclusions that would initially have been quasi-spherical are deformed such that they are very long compared to their width; we can understand this from the fact that the volume of the inclusion is not changed.

Let us assume that an inclusion is rolled such that it forms a long, wide defect, and that the piece of steel containing the defect is manufactured into a component where strength is critical, such as a con rod. If the 'long side' of the defect is aligned with the applied stress, then it presents little cross-sectional area, and its effect in terms of stress intensity factor means it is less likely to succumb to sudden failure from an overload. If the 'long side' of the defect, or largest cross-sectional area, is perpendicular to the applied stress then the tensile and fatigue strength of the material is much lower, and it is also more susceptible to the opening of an internal crack and failure from a single overload.

Fracture mechanics tells us that the product of fracture stress and the square root of flaw length is constant for a given material and type of loading. Therefore, by rolling an inclusion flat and loading the material in a direction perpendicular to the rolling direction, we lower the strength of the material.

This is why we have to be careful to align the material from which we make critical components carefully with the direction of stress, where significant anisotropy occurs. Many materials publications will give data for materials in both longitudinal and transverse directions.

Given that loading conditions are seldom simple enough to ensure loading parallel to the long side of inclusions, some companies have taken steps to minimise anisotropy. The main thrust of this work has been to reduce the general amount and size of any inclusions before any processing that would otherwise induce directional differences in mechanical properties.

Ovako has been instrumental in this using its IQ process*. This involves not only producing a steel with lower overall levels of 'problem' elements, such as sulphur, but also in the melting method and shielding of the molten material. This has helped eliminate all but the smallest (sub-20 micron) inclusions, and materials thus produced show improved fatigue properties. In producing materials with a smaller number of inclusions, and by taking measures to limit the size of such defects, gives an improved material after rolling, and there is much less difference in longitudinal and transverse properties.

Be very careful though when buying steels with an IQ reference. Some other general engineering steels have an IQ reference which has an altogether different meaning. So-called Inclusion Modified Steels marketed under the 'IQ' brand by some general steel stockholders aim to improve machining properties by adding extra quantities of elements that can and do cause real problems, as they can contain huge inclusions which, after rolling, run to many metres long in the bar. These render the material useless for highly stressed applications. One engine manufacturer had to scrap a large batch of cams due to such material being used, and I am sure many other components are compromised by the use of such steel.

*Ölund, P., "The IQ-process - the Ovako isotropic quality process", ISSN 0284-3366

Fig. 1 - Billet con rods need to be aligned to the rolled structure of the material to take advantage of the maximum strength of the material

Written by Wayne Ward

For a number of years there have been rapid prototype parts run on racecars on everything from Formula Student to Formula One. The method offers a way to have small numbers of complex parts made quickly and in a cost-effective manner, with the costs of design changes in many cases being restricted only to those relating to the design itself.

There are many polymer materials that can be used not only for ancillary parts on engines (covers for example), but in manufacturing (wax cores for investment castings) and engine development rigs (transparent covers for rigs studying oil motion, for example). However, the applications of rapid prototyping using the available polymer materials are limited in race engines owing to stress and temperature constraints.

For a number of years there have been metallic materials used for rapid prototyping, and the technology is now producing truly impressive parts with excellent mechanical properties. A RET-Monitor article (www.ret-monitor.com/articles/898/advanced-materials) looked at some of the available materials, and previously one of them, the aluminium alloy Al-Si10-Mg, was singled out for brief examination (see www.ret-monitor.com/articles/828/new-application-of-aluminium-alloy/). The list of available metals is expanding and already encompasses a number of materials of interest to the engine designer, including aluminium alloys, high-strength steels, titanium and high-temperature materials.

An application for which rapid prototyping is finding increased use in racing, in the field of metals processing, is in the production of sand castings, where it is now possible to 'print' sand cores and outer moulds directly for major structural castings such as blocks and heads. In one such method, the rapid prototyping machine produces a layer of sand that is subsequently and selectively treated with a binding agent in the relevant areas. In another, sand that is pre-coated with resin is fused using laser sintering. At the end of the process, the untreated sand can be easily removed to reveal the solid sand tools.

Owing to the fact that cores produced by such methods do not need to be withdrawn from core boxes, it is possible to model castings such that the cores need no draft angle, and they can be more complex than ever before, with many kinds of finely detailed internal features in the casting now possible. A secondary benefit is that your casting CAD model needs far less post-processing at the foundry than before.

The method is still developing, and it is not at the stage yet where it can economically replace traditional methods of producing castings for anything other than very short runs of prototype parts. However, an important advantage is the rapidity with which the first cast parts can be produced using these methods.

Owing to the fact that production of core boxes and machining of moulds is no longer required, the production times can be impressively short. Of course, none of this precludes the requirement for careful qualification of the castings to ensure integrity and any necessary casting development.

Fig. 1 - Rapid prototyping of casting tooling makes production of complex structural parts, such as this racing cylinder block, much quicker (Courtesy of Vitesse Engineering Services)

Written by Wayne Ward

Where hollow-stemmed valves are filled with a liquid metal, the main mechanism of heat transfer from the head of the valve may no longer be through the seat, and in these cases the choice of seat material is not of such critical importance thermally.

In general, bespoke race engines will have valve seat inserts fitted and machined in situ. There have been attempts in racing to used sprayed valve seats, where the seat material is plasma-sprayed onto a prepared head and then machined back to the correct form. For small quantities this process is extremely costly, although for large production quantities (as in manufacturing road vehicles) it is increasingly common and is an economical process.

The most common materials used for making valve seats for bespoke race engines are copper alloys, as mentioned in an earlier RET monitor article about copper alloys. They have excellent thermal conductivity, good wear characteristics and reasonably high strength. Copper beryllium alloys that contain small additions of beryllium are commonly chosen for race engines, with those who specialise in seats offering them as 'A3 alloy' or 'A25 alloy' seats, although beryllium-free alternatives such as Moldstar 90 and Moldstar 22 are gaining in popularity in racing. As you might conclude from the name Moldstar, such alloys (and also beryllium copper alloys) are commonly used for manufacturing mould tooling, where conductivity is also a highly desirable property.

So, apart from these materials, what else could we consider? Well, many people persevere with the production valve seat materials for production-based race engines. Cast-iron heads had the seat machined directly into them, so such cast-iron materials can be considered as candidates for seats, as can steels.

White cast-iron inserts, which are very hard, have also been used in production vehicles, as have powder metallurgy steel materials. A number of companies offer these powder metallurgy materials and finished seat inserts for sale. They range from steel 'pre-forms' that are infiltrated with copper alloys, to a tool steel matrix infiltrated with graphite-containing iron alloys.

Beyond this, and into the realm of more exotic materials, some silver alloys are attractive because of their high thermal conductivity, but less so from the point of view of strength and the fact that parts may 'go missing' from stores. With precious metals at a premium at present, and likely to remain so, this is perhaps not the time to be looking at fitting silver-based alloy seats to your engine!

Fig. 1 - Powder metal materials are a possible choice for race engine valve seats

Written by Wayne Ward

Since the widespread adoption of Ti10-2-3 on aircraft such as the Boeing 777 for critical structural parts of the landing gear, another alloy has become the preferred option for newer aircraft, owing to a combination of higher strength and lower cost. This alloy, Ti5-5-5-3, which contains 5% aluminium, 5% molybdenum, 5% vanadium and 3% chromium, was also used on the 777, not in critical structural applications but for other systems such as ducting for the wings' leading edge de-icing systems. It is used on the newer 787 and on several other popular aircraft and is suited to structural applications such as major landing gear components.

The primary advantage of Ti5-5-5-3 compared to Ti10-2-3 is the increased strength which, in the case of the ultimate tensile strength (UTS) of billets, is 4% higher than Ti10-2-3 and 27% higher than Ti-6Al-4V. In forged form, the UTS of the Ti5-5-5-3 material is almost 40% greater than Ti-6Al-4V. The comparison in terms of other properties such as fracture toughness and elongation is not so favourable however - Ti5-5-5-3 has 14% lower fracture toughness and 38% lower elongation than Ti-6Al-4V, but compared to Ti10-2-3 it is better in both respects.

Ti5-5-5-3 was developed from the Russian alloy VT22, the composition difference lying in the fact that the 3% chromium in Ti5-5-5-3 replaces the 1% chromium and the 1% iron in VT22.

Although unlikely to present an advantage to the producer of race engine components, Ti5-5-5-3 can be hardened in larger sections than Ti10-2-3 (150 mm compared to 75 mm). A real advantage of Ti5-5-5-3 compared to Ti10-2-3 lies in its heat treatment, where the solution treatment stage is followed by air cooling rather than water quenching. This makes significant distortion of parts less likely.

A compelling advantage of using Ti5-5-5-3 in place of Ti10-2-3 is cost. Despite the fact that 5-5-5-3 offers a range of mechanical advantages, the raw material is less expensive.

Ti5-5-5-3 also has excellent fatigue behaviour. US patent application 2007/0102073 gives it about a 60% advantage compared to Ti6Al-4V in fatigue testing carried out at room temperature and with a stress ratio (min stress divided by max stress) of 0.1, and other sources state that the fatigue strength of cast Ti 5-5-5-3 is better than that of many wrought Ti alloys.

So for which applications might we consider this material?

Con rods are perhaps the obvious one, but the advantage would depend on the amount of material in a conventional Ti6-4 rod that would offer an advantage if replaced. Where a part is designed with stiffness in mind, as in the case of a con rod, replacing the original material with a stronger one doesn't offer much of an advantage, and the modulus of this material is within the range expected of Ti-6Al-4V. The density of Ti5-5-5-3 is 5.4% greater than that of Ti6Al-4V, however, so it might prove difficult to find an advantage in this application.

Cyclically loaded fasteners, which place a premium on high fatigue strength and low stiffness, would be a good application for this material, as well as possibly some valvetrain applications such as spring retainers. Work done previously on titanium gears in which the parts were made from Ti-6Al-4V would benefit from this newer material, with all salient properties such as bearing strength, compressive strength and fatigue strength being significantly improved.

However, compared to Ti6Al-4V, Ti5-5-5-3 is said to be difficult to machine, with cutting speeds possibly reduced by as much as 50%.

Fig. 1 - The microstructure of Ti 5-5-5-3

Written by Wayne Ward

The Ti-6Al-4V alloy is used widely in motorsport for fasteners, con rods and various other components. In fact it enjoys a virtual monopoly in the manufacture of mainstream titanium race components, with the exception of poppet valves, for which there are a number of alloys used specifically for them.

In the aerospace industry there is a much wider range of titanium alloys used for structural parts such as landing gear, and these alloys would make good candidates for a number of race engine applications, especially con rods. These alloys have disadvantages though, such as higher material costs due to the increased quantity of alloying elements and the lack of availability compared with Ti-6Al-4V. In aerospace, the higher cost can often be justified by the reduction in mass gained from their use, especially where they can replace steel components.

Imbued with greater tensile properties and fatigue strength than Ti-6Al-4V, they offer a real advantage when seeking to minimise mass. In decreasing the mass of a con rod, there are further advantages in that the crankshaft can also be made correspondingly lighter.

Ti 10-2-3 is an alloy containing 10% vanadium, 2% iron and 3% aluminium, and has found widespread use in the aerospace sector for a number of years. In billet form it has a 24% higher yield strength and 22% higher ultimate tensile strength compared with Ti-6Al-4V, and the advantage in these properties is increased in the forged state, with both strength measures being 27% higher compared with Ti-6Al-4V. In terms of fatigue strength, it enjoys a large advantage - according to published data sheets, axial fatigue amplitudes are about 50% greater for Ti 10-2-3.

Compared with Ti-6Al-4V, however, Ti 10-2-3 has lower elongation and fracture toughness in the forged state. Its elastic modulus is also towards the lower end of the range that we would typically find for Ti-6Al-4V.

Of course, for a con rod, strength isn't everything, and designing a con rod with the correct overall stiffness (and the correct stiffness in certain critical areas) plays a large part in the successful application of a new material. There are certainly areas on an optimised race con rod, however, where fatigue strength will be the limiting factor, and alloys such as 10-2-3 should therefore provide an advantage in terms of mass reduction.

In aerospace applications it is used at temperatures of up to 260 C (500 F), so its temperature capabilities are sufficient for all of the applications where Ti-6Al-4V is currently used.

Fig. 1 - The Boeing 777 is reported to use Ti 10-2-3 landing gear parts, which were formerly manufactured from steel (Courtesy of Boeing)

Written by Wayne Ward

Besides pistons, I mentioned a number of other applications where aluminium-beryllium might be considered, including static applications. In this article we will look briefly at some of the other materials used in reciprocating applications where aluminium-beryllium is now banned.

The most significant potential application of aluminium-beryllium, however, is definitely the piston, and in terms of low-mass alloys there are clearly a number which are currently widely applied. Aluminium alloys are the clear favourite here, and there are a few that are most commonly used for forged racing pistons.

There have been attempts to replace these over the years, yet for most people the old favourites remain the best choice. But there are a few significant developments in aluminium alloy technology that have been seriously considered for pistons and could be used for other reciprocating applications.

The first is a recently developed processing method that involves extremely rapid cooling from the molten state of specially developed alloys. The resulting material structure imbues them with great strength and significantly improved stiffness. The problem with these alloys in the past though has been a lack of ductility, but developments have improved this situation recently.

Lithium has lower density than aluminium and is used as an alloying element in a small but growing number of alloys, mainly for aerospace applications. In trials as piston alloys, however, they have produced with mixed results.

The combination of lower density and increased stiffness makes them an attractive prospect for pistons and other reciprocating parts. For instance, there are alloys that offer an improvement in modulus (a measure of stiffness) of 16% while being 10% less dense than a typical aluminium alloy. More commonly available Al-Li alloys offer a small density advantage (<5%) and an increase in stiffness of 5-10%.

Metal-matrix composites have been discussed in RET Monitor previously for pistons, and these materials are in the same position as aluminium-beryllium in that they are banned for Formula One applications. But they are used successfully outside Formula One for pistons and other applications, and there would therefore seem to be no logical reason to ban them from Formula One, which is often perceived to be the one series in motorsport which is a technical showcase.

Magnesium, again banned in Formula One, has been covered in RET-Monitor in an advanced metals feature. It has been used successfully for pistons in the past, and has been the subject of trials and testing relatively recently for more highly optimised machinery.

There are a number of good reasons why the material isn't used for pistons, but the lack of thermal conductivity and fact that it is so reactive are two that are of importance. Before being banned, however, it did find use in non-piston applications.

Fig. 1 - This two-stroke racing piston is make from a rapidly solidified aluminium alloy, giving exceptional strength (Courtesy of RSP Technology)

Written by Wayne Ward

In terms of four-stroke con rods, most of us will be familiar with the concept of using a bushed small end. In the vast majority of cases this will be a bronze material, although for higher pressure applications copper-beryllium has also been used with success. There have been occasions where coated high-strength steel has been used, but this is uncommon and bronze alloys continue to be the most popular.

Continuing with bearings, the backing material of bearing shells is still quite often a bronze alloy, although steel is often used now. While not as stiff as steel, the bronze material does have better thermal conductivity.

As we might imagine, most uses for copper are as a bearing alloy, and while the previous two applications are rotating bearings, the next are sliding bearings. In pushrod applications, it is common to machine the block in order to sleeve the lifter bores, and these sleeves are commonly bronze materials. The capabilities of copper alloys as bearing materials are especially valued where keyway lifters are used.

Copper alloys are used extensively as valve guides. This is not only for the bearing characteristics of the materials being used but also thermal conductivity. There are other materials that could be suitable candidates for use as a valve guide, but the various copper alloys in this application offer good wear characteristics and the ability to conduct heat away to the head, allowing the stem seals to operate at a lower temperature and therefore with greater reliability. Copper-nickel-silicon alloys are often used in this application.

For the same reasons, copper alloys are favoured materials for machining valve seat inserts. Here, they need to have good wear characteristics, but also good thermal conductivity as they conduct most of the heat away from the valve and into the head. There are a number of good candidate alloys for this, although many now seem to gravitate towards copper-beryllium for this purpose, often with different alloys being used for inlet and exhaust seats.

In sealing the cylinder liner to cylinder head interface, people often use flat sealing rings, and these are sometimes referred to - mistakenly - as beryllium rings. These have been made from beryllium-copper alloys, although this can be an expensive route to producing these parts, which can perform well in a number of much cheaper alloys, and a number of relatively cheap bronze materials are used successfully.

New alloys with improvements in thermal conductivity and strength are available and we will surely see these alloys being considered for racing engine use in years to come.

Fig. 1 - Cylinder head fitted with copper-beryllium valve seats

Written by Wayne Ward

In 2000, the editor of Race Engine Technology Ian Bamsey wrote in Racer, "Since 1998, Ilmor has manufactured pistons from an aluminium-beryllium alloy, thereby reducing their weight by a third, possibly more, and gaining enhanced thermal conductivity. The cost of this alloy, and the fact that fine beryllium dust particles arguably constitute a health hazard, has led to an effective ban on its use, imposed by the FIA. Under pressure from McLaren and Mercedes, however, this ruling, for which Ferrari lobbied hard, has been postponed to the end of the current season."

In granular form beryllium is indeed hazardous to health. There is a disease associated with it called berylliosis and this is a chronic condition affecting the lungs. Beryllium is used widely in racing engines, in copper-beryllium alloys, and herein lies the key to its safe use. The beryllium in valve seats is alloyed with copper, so it is soluble in copper and is not simply embedded particles of beryllium within the copper matrix. On the other hand, aluminium-beryllium is a metal-matrix composite, with finely divided beryllium reinforcement held in an aluminium matrix, so it is more hazardous.

But I have heard a rumour from more than one source that a certain team objected to the use of this material after it was refused an exclusive supply of it for its own use. Only then, I'm told, did the team's objections on health and cost grounds become so strong. Ron Dennis, whose McLaren team benefited from using aluminium-beryllium pistons from 1998 to 2000, insisted that all of the health risk lay in the manufacture of the parts, not in their use. The material is still used in many applications to this day, albeit outside motorsport.

The reason this material proved to be ideal as a piston material lies in its mechanical properties when compared with those of standard materials. Beryllium has a very low density of 1.85 g/cm3. With an atomic number of four, it is the second metal (if one discounts hydrogen) in the Periodic Table after lithium. It is also very stiff, with an elastic tensile modulus in excess of 280 Gigapascals (GPa). By comparison a typical aluminium alloy has a density of 2.7 g/cm3 and a modulus of 72GPa.

The specific modulus (modulus divided by density) for beryllium is almost 470% greater than that of aluminium. Beryllium's thermal conductivity is also greater than that of a popular aluminium piston alloy by a large margin. So adding beryllium to an aluminium alloy would have some very positive effects. A typical commercially available aluminium-beryllium alloy has a specific stiffness 250% greater than a good piston alloy, and 44% greater thermal conductivity.

The one downside that has been described to me is that, when pistons fail in service, there is very little chance of the source of the failure being proven. Such is the material's lack of ductility that it is said to give the appearance of having shattered.

So where else might we have expected to find a material with such properties? Cylinder liners are a good application to look at here. At the time, the linerless block had yet to become the norm in Formula One, and both aluminium and steel liners were in use. Aluminium-beryllium would surely have supplanted both, giving significant advantages in stiffness and mass, and it would certainly have been the material of choice for some valvetrain components in the universally used (at least in Formula One) pneumatic valve return systems. There are certainly other applications where we might have seen it used, had it not been outlawed.

Outside motorsport, the most popular applications of these materials are in satellites, where low launch mass is important, and in large optical instruments such as reflecting telescopes - the superb image here of Jupiter was taken by a space-based telescope. One such instrument, the James Webb Space Telescope, is scheduled for launch in 2014 and is technologically possible because of materials such as aluminium-beryllium. Its aim is to study the very earliest galaxies formed in the Universe, which rather puts race teams squabbling over materials into context.

Fig. 1 - Satellites and optical instruments are popular applications of Al-Be alloys. Images like this are, in part, thanks to their use

Written by Wayne Ward

There is now, however, a wide range of materials which can be processed by this particular method. If we consider the aluminium material to be advanced because of the opportunities that it offers us to make previously impossible forms, then we should also consider the other materials which can be processed by these means to be advanced.

Certainly one which might be of most interest to us as race engine designers and engineers is Titanium. As we might reasonably expect, the titanium alloy which is available is Ti-6Al-4V. Owing to the fact that the stock from which the metallic powder is made is available at reasonable cost and, most importantly, of good quality, it is natural that this material has been chosen over some of the other available alloys. Also, if people are considering replacing titanium parts with ones made by this new process, there is no risk from switching to an unknown alloy. The tensile properties of the alloy are equally as good as the wrought alloy given that the process was in its infancy only a few years ago, this represents great progress. More remarkably, the available fatigue data for this material shows that it is comparable to the wrought material.

In addition to aluminium and titanium, there are other materials which might be of interest. The main target market for this process is the aerospace industry, and so it is no surprise that we find two materials available which we would normally term superalloys intended for high-temperature use. There are Cobalt-Chromium-Molybdenum alloys and one of the Inconel alloys is also available. Whilst cobalt-chrome might not be a material that we are familiar with, there are a number of applications of inconel alloys for racing purposes. Certainly, as we have mentioned elsewhere, Inconel alloys are widely used for exhausts and at a recent trade show, experimental exhaust parts were shown by EOS (see accompanying picture) as a demonstration of the possibilities that the technology offers. Such was the quality of the part in question that one had to examine it carefully to see that it wasn't made by the normal method, e.g. fabrication.

Two precipitation-hardening stainless steels are available, and again, these don't seem to find widespread use on racing engines.

One of the maraging steel alloys is among the choice of alloys and this may be of interest to some of our readers. Maraging steels differ from most steels in that in their as-quenched state, they are soft and require a low temperature aging process in order to achieve their final hardened state. In this regard we might think of the hardening process as being similar to that of aluminium alloys.

This new manufacturing process turns some widely-available and normally unremarkable materials into advanced materials, and clearly the wide range of useful materials offers us great possibilities for making novel engine components.

Fig. 1 - The experimental exhaust part was made by laser sintering of metal powders (Courtesy of EOS / 3TRPD)

Written by Wayne Ward

We might consider a mundane polymer not to be an advanced material. If we take two thin sheets of it, and separate these with a honeycomb structure, we create something very stiff and lightweight. Some may consider this to be advanced; however, others would correctly point out that much better materials of the same type are available. However, by special processing, we can fill the void in between the two sheets with a hard foam of the same material. Again we have a stiff, lightweight structure, but without the complication of manufacturing a honeycomb structure and without the risk of delamination of the skins from the honeycomb. Bonding to polymers is often problematic and therefore risky and so we might well contend that this new structure, made entirely from a mundane polymer is an advanced material, as it offers us new possibilities in design. Such materials don't find use in racing engines, and probably not in racing cars, although they have existed for many years.

I would certainly consider the material to be advanced in the above-mentioned state, and it is on this basis that I propose the equally mundane aluminium alloy AlSi10Mg to be advanced. Owing purely to the fact that it is, currently, one of the only available aluminium alloys capable of being processed successfully by 'direct' rapid prototyping methods, it offers unique opportunities to us. The material itself, in terms of composition is not a new one, nor is it remarkable in terms of its mechanical or physical properties. Possibly it has been chosen for this application owing to the fact that it is available as a 'master' casting alloy and so will contain extremely low levels of impurities.

The process itself uses very finely divided 'powdered' material which is fused together to form a complete part by the use of lasers. The quality of the powder is critical to the success of the process, which has only recently been available. I was fortunate enough to have some parts made by the same process in the USA a few years ago, but the material lacked density, was extremely weak and the metallurgical structure, even in areas of the part with reasonably density was not far short of abysmal. However, the new process offers a good level of strength, commensurate with that which we might expect from a casting in the same material. We are now, however, bound only by our imagination and knowledge of how the process works in designing parts which were, hitherto, impossible to make, as shown by the accompanying picture. This complex part is made as a single component.

Fig. 1 - This complex aerospace part is made as a single component.

Written by Wayne Ward

Magnesium enjoys a very low density and so is naturally attractive to the designer of a bespoke engine. With a density of 1800kg per cubic metre, it is less than twice as dense as water and is over 30% less dense than a typical aluminium alloy. However, in keeping with most metals, its stiffness or, more accurately, elastic modulus is roughly in proportion to its density and this is typically around 45GPa. However, providing that we have enough space, and that the part isn't stressed purely in tension or compression, we can take advantage of the low density to employ the same mass, intelligently placed, to provide greater stiffness than the equivalent mass of aluminium or steel. The fatigue strength of magnesium can though compare favourably to aluminium and work is being done to develop new alloys with improved tensile and fatigue properties.

So, where have we seen magnesium used in engines? Well the obvious application is in castings. Every major casting of a racing engine has been made at one time or another from magnesium. There have been experimental Formula One cylinder blocks (at a time when this common material used widely on series production engines was legal in Formula One), cylinder heads, crankcases, cam carriers and covers, not to mention the many non-structural covers and castings around the engine. So, apart from nonsense rules, why is it not more widely used today? Maintenance and corrosion are two good reasons. Common complaints about magnesium parts are that they are prone to threads being pulled out (people tend to make a simple material substitution and forget that threads will pull with maybe 30% less applied stress) and that the parts corrode very quickly. People who manage to keep magnesium and moisture apart are generally much happier to use the material. It is among the least noble metals and is prone to general corrosion and particularly susceptible to galvanic corrosion. This is why magnesium is most commonly seen with some sort of surface treatment applied. It has been used as a piston material and owing to good high temperature mechanical properties it might appear a good candidate at first examination. However, it has some other properties which render it less attractive as a piston material. Among these is low thermal conductivity. It has been used in valvetrain applications with success too. Anywhere where there is little or no moisture, and an oil film seem to be relatively trouble-free applications.

BMW has recently introduced a series production magnesium composite cylinder block, where the outer part is magnesium and the inner part, which is in contact with the water circuit is made from aluminium. This is one of the major reasons for the large decrease in mass compared to the previous equivalent engine from the same manufacturer. Magnesium has been popular for none-structural covers on motorcycle engines for decades and in the push for smaller lighter engines as we seek to improve fuel economy, it will become a more common material for series-production motor vehicles. The accompanying picture shows a 1952 Norton motorcycle engine which uses magnesium crankcases. Almost 60 years later Formula One rule makers disallow this material, whilst it becomes common on road cars. In racing, we must be careful not to become an irrelevant backwater of old materials technology.

Fig. 1 - 1952 Norton engine.

Written by Wayne Ward

Metal matrix composite pistons are not a new concept, but haven’t been widely taken up. The natural arena for development of new piston alloys would have been Formula One, but this avenue of development has been closed for the time being. The ‘trickledown’ of technology to those formulae without the large budgets that are the preserve of Formula One is an important part of the progress which has been made in racing engine design. However, even without this helping hand, MMC materials have continued to be developed and applied to engines.

Recently, Race Engine Technology magazine’s Wayne Ward discussed the subject of MMC piston materials with Dr. Jonathan Silk from metal matrix composite specialists AMC-MMC in England. The company has more than one piston alloy available, but where technical regulations do not specifically prohibit it, AMC225XE is their material of choice. This contains small particulate reinforcement in the form of ceramic particles of silicon carbide. Silk reports that the particulate reinforcement particles are very small (typically 3 microns) and this means that the alloy can be forged without difficulty to ‘near net shape’ and can be machined with similar ease to a high level of surface finish. The accompanying pictures, supplied courtesy of AMC-MMC, shows a sectioned motorsport piston forging in 225XE of the popular box-bridge type with stiffening ribs which tie together the two piston pin bosses. A further picture shows the microstructure of the material with the silicon carbide particles within the aluminium alloy matrix.

As can be seen from the table of properties presented in the aforementioned RET Monitor article, the mechanical properties of the material are very impressive, and it ticks many of the boxes that we would ask for in terms of an improved piston alloy, being stiffer and less dense than the current favourite piston alloy, 2618. It also has good thermal conductivity which is something else which is valued by piston designers. Silk states that AMC225XE is able to offer greater fatigue strength than 2618 at all temperatures.

Whilst AMC225XE is banned under the current Formula One engine regulations, it is designed to have a specific stiffness which just comes within the chassis limit of 40 GPa /(g/cc) and so still finds widespread use in Formula One and other areas of motorsport. Thankfully this means that 225XE remains available to racing engine manufacturers outside of the Formula One regulations. Dr Silk stated that the material is currently used for a range of piston applications and also that further developments of these alloys continue which should yield materials of even greater benefit to the piston designer.

Written by Wayne Ward.

The picture which accompanied the aforementioned article showed additional counterweighting mass in the form of cylinders or slugs of tungsten pressed into place in each counterweight. There are some advantages to this method, one of which is that the strength of the tungsten material is not too critical. There are several disadvantages of this method, probably chief among them being the small amount of tungsten which can thus be added. Also, the optimum place to add such heavy metal is at the very outside radius of the counterweight. Clearly, in order to contain these inserts, we require an amount of steel to surround in order to retain these dangerous inserts in place. There are serious consequences if one of these parts should become detached from the crankshaft - this generally happens at high engine speeds, and the high speed tungsten is easily capable of punching a hole through the side of the engine and any other objects blocking its bid for freedom.

To place the tungsten at the very outermost position, we must look to fasten these to the crankshaft, generally using threaded fasteners. If one works through the necessary calculations to determine the size and the strength of the fasteners required for this critical job, you soon come to realise that the strength of the tungsten itself is also critical. There are a number of material stockholders who supply tungsten alloys to the racing industry; it is used extensively as chassis ballast in Formula One. Owing to the fact that the loads due to these ballast parts is small in comparison with their dimensions owing to low levels of acceleration, there is scope to use a larger number of less critical fasteners and therefore the design engineer can tolerate a low strength tungsten material. Popular materials for chassis ballast typically have a yield strength of 550MPa.

For optimal counterweighting however, we need strong tungsten materials, because we naturally want to use the minimum number of fasteners for retention. The purpose-designed fasteners have a much lower density (approximately 50% lower) than the tungsten alloy, and each counterbore or clearance hole provided for these fasteners has a density approximately one fifteen-thousandth of the density of the tungsten, because it contains only air. Therefore any extra fasteners taking up volume which could be occupied by tungsten is diluting the effectiveness of the counterweighting.

In order to produce high-strength dense metal materials we must look to work-hardening methods. The main application for high-strength heavy metal materials is armour-piercing ballistics, and it is therefore to companies supplying the military that we must turn to gain a supply of this specialist material. In Europe there are a small number of suppliers who can supply material with a yield strength exceeding 1000MPa, and one in particular has two ranges of round bar which are termed ‘medium-calibre and large-calibre’, in reference to their normal application.

Written by Wayne Ward.

Whilst in some applications, a low modulus material is very useful (fasteners for example can benefit from having a low elastic modulus), in many of the applications where aluminium is used, high stiffness would be a useful benefit to the designer. Most aluminium alloys have a modulus of around 70 GPa (10,000,000 psi), which is approximately one third of the elastic modulus of steel. Without resort to the addition of particulates to the material i.e. powder reinforced metal-matrix composites, there are commercially available materials used in other industries which offer a substantial advantage over conventional aluminium alloys in terms of stiffness.

Again, some engineers (Formula One again) are limited by the ‘glass walls’ imposed by regulations. The regulation which applies here is that limiting specific modulus of elasticity to be less than 40 GPa / (g/cc). This specific modulus is the ratio of elastic modulus to density. We would typically be comfortably within this limit. As examples, 7075 and 6061 both have a specific modulus of 25.5 GPa / (g/cc), and this is typical for conventional alloys.

By specifically increasing the content of certain elements, it is possible to create aluminium alloys with a specific stiffness far exceeding those typical values. The methods of processing these special alloys are very different to those generally used, with incredibly high rates of cooling used to promote a very unusual material microstructure. One particular example produced in Europe, which is commercially available to all of us, has a specific stiffness of 53.3; this being comfortably more than twice the typical value. However, in Formula One, engineers cannot use this. There are though several alloys which fall below the mandated limit which would be of interest to Formula One engine design engineers, some of which are also blessed with high-temperature capabilities, making them possible replacements for alloys such as 2618 in some applications.

So, what are the disadvantages of these alloys in terms of their mechanical properties? Many of them suffer from low ductility. Whilst ductility is a property associated with plastic deformation, it is of significance to those of us aiming to design parts which we believe to operate in the elastic region of the load deformation curve. Where there is contact, we often require the material to have a certain amount of ductility to allow the contact to conform slightly, especially at the edges of highly-loaded contact regions. Where ductility is too low, lack of ductility can lead to cracking and consequently low fatigue life. If we aim to introduce these alloys, we have to design our contact areas pretty carefully. As an example, alloys of this type often have less than 3% elongation, and in some cases around 1%. Sometimes, although not always, these materials can suffer from low tensile and fatigue strength. However, some of these materials display high tensile strength and exceptional fatigue strength compared to standard alloys. Experimental very highly alloyed aluminium alloys produced only in extremely small quantities, by the same methods as the high-stiffness alloys above, have recorded fracture strengths approaching 1500 MPa (215 ksi)i

There are many applications within the racing engine where these alloys could be employed and they will doubtless become more popular as time goes on. The only limits at the moment are the availability of materials in sufficient quantities or in suitable sizes and the price, which is high in comparison with conventionally-processed alloys.

Written by Wayne Ward.

It is rumoured that at least one manufacturer has seriously investigated titanium as a crankshaft material in Formula One, although this was some years ago, and I’m not aware of anybody looking at this currently. The steel crankshafts currently used in Formula One are highly optimised and to change to a material with such different mechanical characteristics would require big changes. Certainly we could expect crankpin diameters to be increased, and therefore the big end of the con rod would be consequently larger. This would have an impact on the crankshaft axis height from the base of the engine, requiring a deeper lower crankcase (sump), and therefore having an effect on the packaging of the engine in the car. Given the pressure that has been placed on the engine suppliers to make engines with the lowest possible crankshaft centreline height, it is doubtful that anyone would be tempted to spend any significant effort on this development.

Formula One regulations currently proscribe piston pins in materials other than an ‘iron based alloy’:

5.14.2 Piston pins must be manufactured from an iron based alloy and must be machined from a single piece of material.

However, there have been several past attempts to use titanium alloys in this application in Formula One, with varying degrees of success. I have spoken with an engineer who worked on one of the V10 engines which had hollow titanium piston pins with longitudinal fibre reinforcement. The cost of these parts was described as ‘extremely expensive’, with the material cost being a very significant reason for this expense. The material in question was produced specifically for this application.

A more pragmatic approach, and one that it certainly less expensive, is to take a commercially available titanium alloy, and machine piston pins from this, taking advantage of the lower density to produce a part of equivalent stiffness but with lower mass. The stiffness we are talking of here is the ‘crushing’ or ‘ovalisation’ stiffness, and we might size the pin to have a maximum design deflection under a given load. In a very trivial example, in maintaining the same outside diameter, the theoretical mass saving by substituting steel for titanium is in the region of 35%.

This approach of machining parts from commercially available bar stock has been followed by Formula One engine manufacturers with more limited development budgets in the past, and even some people outside of Formula One.

However, it is not simply a case of matching stiffness. Strength, and particularly in respect of fatigue, has to be carefully assessed. In this regard steels can be surface hardened to introduce significant compressive residual stress which helps to increase fatigue life. Titanium is not able to take advantage of the same levels of residual compressive stress, and therefore pins need to be of larger section than those that result from calculations to match stiffness.

If we aim to use titanium in this contact application, we need to pay special attention to surface coatings also, as titanium has notoriously poor contact behaviour.

Written by Wayne Ward.

The main applications for titanium in racing engines currently are connecting rods, valves and fasteners. Indeed, there are a number of production engines where titanium is now the material of choice for connecting rods and valves. Whilst the first production engine use of titanium was on motorcycles in 1988 (on the Honda VFR750R), the high-profile introduction of this to road-cars followed shortly afterwards, again by Honda with their NSX sports car. More recently the Corvette has followed this trend with the Z06 using titanium con rods.

If we restrict ourselves to pure racing engines, by far the most popular material for con rods is Ti-6Al-4V, an alloy containing, as one might reasonably expect, 6% of aluminium and 4% of vanadium. This is the predominant material for all uses available on the world market – it is produced by many of the leading materials producers and by dint of the amount of material produced is relatively reasonably priced. It can be heat-treated to high levels of strength and is generally of a high quality, although I have seen absolutely appalling quality material in this grade – beware those offering bargain-basement prices for titanium if you are unable to check the quality of the bar-stock beforehand.

Besides Ti-6Al-4V (also known as Grade 5 Ti or TA46), there have been various grades examined for con rods. The beta forging alloys have been used for forged titanium racing rods, particularly Ti-550 and Ti-551. Both of these alloys have additions of Aluminium, Molybdenum, Tin and Silicon, with 551 being more highly alloyed and offering higher room temperature strength. However, after speaking to a number of well-known con rod manufacturers, it seems that a lot titanium con rods are no longer made from forgings, but from blanks made from wrought materials (plate in particular).

As far as alternatives are concerned, there are a few materials which look promising, and some more that are being developed for use as racing rods. Some of the less common materials look to have excellent mechanical properties and good fatigue behaviour. However, unless there can be guaranteed continuity of supply and quality, people are either forced to stick with Ti-6Al-4V or to purchase sufficient quantities of the material to cope with fluctuations in the supply of the raw materials.

Perhaps where the rules allow it, the titanium aluminide materials would be a good replacement for the conventional titanium materials. It has been confirmed to me that this material has indeed been used successfully for con rods in racing applications, although, in this particular application, it has been shelved owing to high material costs.

It is outlawed in Formula One, under regulation 5.13.1: Unless explicitly permitted for a specific engine component, the following materials may not be used anywhere on the engine :

a) Magnesium based alloys ;
b) Metal Matrix Composites (MMC’s) ;
c) Intermetallic materials ;
d) Alloys containing more than 5% by weight of Beryllium, Iridium or Rhenium.

Where intermetallic materials are defined by regulation 5.12.3: Intermetallic Materials (e.g. TiAl, NiAl, FeAl, Cu3Au, NiCo) – These are materials where the material is based upon intermetallic phases, i.e. the matrix of the material consists of greater then 50%v/v intermetallic phase(s).

There have been development efforts by a number of Formula One engine manufacturers in the direction of increasing the proportion of titanium aluminide to bring it close to this 50% limit – another example of the Formula One rules not working as intended. As with titanium now finding its way into production engines, it may be that, in years to come, titanium aluminide will become the material of choice. Rules which ban it simply stifle development of racing and production car engines. If you have open rules, the expenditure will be self-governing, with the titanium aluminide rod application as an exemplar.

MMC titanium alloys have also been developed for motorsport applications with additions of ceramic particles for additional stiffness.

Written by Wayne Ward.

Whilst Formula One is frozen at present in terms of introduction of new materials owing to the homologation

of the engines, we should not fool ourselves into thinking that the materials departments of these organisations are asleep or redundant. There is much to be gained from being prepared for any new regulation change, and the Formula One engine manufacturers will be very busy testing and evaluating improved materials.

The general trend is always for an increase in a certain mechanical or physical property of the material, and with the increased number of races that a Formula One engine must now run before being replaced without penalty, increased fatigue resistance is probably now more important than ever.

For structural castings over the years, we have seen the foundries pushed to move from the old LM alloys such as LM25 (a 7% silicon alloy which has an aerospace equivalent called L99, also equivalent to A356) to higher strength alloys such as L169 (equivalent to A357), which is distinguished by its higher magnesium content. Other A35x specification alloys have also found widespread use in Formula One and some other formulae, owing to better fatigue resistance, but there is a concerted push by some to move to more exotic aluminium casting alloys which promise outstanding fatigue behaviour. These exotic alloys have particular difficulties in producing the quality of castings required but, once casting development is completed, they promise large gains, so we can expect their use to become widespread as time goes on, especially under new engine regulations.

There have been many attempts to improve upon the traditional piston alloys over the years, but still many keep returning to the old favourite RR58 / 2618A or something only slightly modified from this specification. RR58, containing 1.9 – 2.7% of copper, has a good combination of properties that have seen it used from World War II (it was developed for the Rolls-Royce Merlin engine) until the current Formula One era. Not only does it have good hot-strength, but it maintains excellent fatigue behaviour and ductility making it resistant to minor damage from debris and even unexpected collisions with valves. I have seen a piston with a failed valve-head embedded in it without causing a catastrophic piston failure. The high-silicon 4032 alloy has also been used for Formula One pistons in recent years. However, these traditional materials should not expect that their place at the top table is guaranteed for much longer. Manufacturers have for many years looked to find something with better properties and now some feel that they are close to finding something considerably better. I have seen results of trials of pistons run in Formula One engines which look very promising. The alloy in question has excellent strength and exceptional fatigue resistance, but lacks slightly in ductility meaning that it has very little tolerance to minor damage, but also any contacts have to be very carefully designed indeed. Whilst this material is probably a little too risky to be introduced at present, work is ongoing to find something with a little more elongation.

I have been in contact with an engine materials expert recently who has access to a new piston alloy which is claimed to be a big step forward from the traditional alloys. We can certainly expect to see this undergoing trials in the near future with a view to introducing this under new Formula One engine regulations. Those who are unrestricted in this way (MotoGP and other formulae such as ALMS) might do well to avail themselves of this material. With much increased fatigue resistance, it may well be that rarest of commodities – a money-saver!

Written by Wayne Ward.

on the market which fall foul of this rule, and not all are produced by the route of ceramic reinforcement which was talked about in last month’s article on the subject. Certainly there are a number of companies producing aluminium alloys by novel methods which have a modulus much greater than the 70GPa that we would generally expect for such materials.

These high stiffness aluminium materials have found applications for both highly-loaded components such as pistons and for structural components to stiffen the entire engine structure. Increasing engine stiffness always seemed to be a popular idea with the chassis designers and a simple material substitution provided a useful increase in overall chassis stiffness as a result. The downside of the very high stiffness materials seemed to be very low elongation resulting in some expensive piston failures. However, newer materials are being developed which seek to address this problem, but Formula One is no longer in a position to help with materials development in this area.

I would like briefly to return to the subject of MMC materials, and specifically the ceramic reinforced metals. The illegality of the ceramic particulate reinforced materials is covered by rule 5.12.5 - a particular clause in the regulations which seems open to question:

“Metal Matrix Composites (MMC’s) – These are composite materials with a metallic matrix containing a phase of greater than 2%v/v which is not soluble in the liquid phase of the metallic matrix.”

I questioned this rule at the time with the team that I worked for, and an eminent professor of materials thought that this rule was written in a way that would allow a lot of MMCs to be used. There are specific classes of MMC materials that would definitely pass this rule and perhaps some of these are being used presently although I have heard nothing to make me think that they are. However, engine manufacturers are notoriously secretive about any competitive advantage they might have!

However, there were some very interesting material developments which did at least make it as far as engine testing if not into races. Without going into too much detail, I can say that carbon-fibre reinforced plastic (CFRP) connecting rods have been tried by at least one engine manufacturer in the past. My understanding is that these were pretty unsuccessful, but there is some anecdotal evidence that they were run at a race meeting in practice, if not in a race.

MMC materials in the form of ceramic reinforced steels were tested very successfully as gudgeon pins within the specific stiffness rule. There is also some evidence to suggest that MMC titanium has been successfully tested and possibly raced before the specific modulus rule was imposed. There has been an unsuccessful attempt in the past to run an aluminium MMC inlet valve in an Formula One engine, but this was a step too far! This is certainly not an exhaustive list of components tested and raced using novel materials, but does give a flavour of what went on when engineers and materials scientists were given a free rein.

For rotating and reciprocating components in an engine there is great merit in reducing component mass. This leads not only to lower loads but also better control of the valvetrain for instance. Whilst there are commercially available low-density tool steels (up to 10% less dense than conventional steels), these weren’t developed with the express aim of low density, but were exploited for highly-stressed engine components. Some of these also possess slightly increased modulus compared to general steel materials.

It is my hope that when the FIA release the new engine rules, that they allow materials development to resume (within a cost cap if necessary). Racing and Formula One has been at the forefront of development of materials and processes which are now used on passenger vehicles, and the benefits of a very light engine are not to be ignored in the field of road vehicle design.

Written by Wayne Ward.

While an MMC can be composed of two metals, most MMC's are combinations of a metal and a ceramic. The metal provides the monolithic, continuous structure into which the reinforcement particles are embedded, and provide a compliant support for the reinforcing particles. Some of the most common MMC's are various alloys of Aluminium (Al) reinforced with particles of Silicon Carbide (SiC).This article spotlights the properties of one specific MMC (AMC-225xe) and some of its varied uses in motor sports. These uses include pistons, cylinder liners, con rods, rocker arms, valve spring retainers and suspension uprights.AMC-225xe is 75% (volume) high-strength Aluminium-Copper alloy (AA-2124) and 25% Silicon Carbide. It's manufacture is based in powder metallurgy techniques, and includes (a) steps to produce ultra fine (2-3 micron) particles of the metal and the ceramic components, (b) the proprietary high-energy mixing process which assures an extremely homogeneous distribution of the components, (c) the HIP-based compaction of the mixture into billets, (d) the forming of the final product (forging, rolling, extrusion), and (e) the appropriate heat treatments.Table One shows the physical properties of AMC 225xe compared to certain other high performance materials. (AlBeMet-162 (62% Beryllium), which was outlawed for political reasons several years ago, is included in the table to illustrate the extremes of MMC technology.)Specific Modulus, one of the most definitive properties, is a non-dimensional number obtained by dividing the Young's Modulus (stiffness) by the density. 225xe has essentially the same density as aluminium alloys, yet is stiffer than Titanium 6AL4V. That is reflected by the Specific Modulus, which is about 60% greater than the aluminium, titanium and steel samples, and interestingly, just under the Formula One regulatory maximum of 40.0.The strength and fatigue performance of this material is also noteworthy, with a room temperature tensile strength 2.25 times greater than 2618 (a popular piston alloy). One company has done its own independent rotating-beam fatigue tests, which show that at 21°C and 150°C, the 10-million cycle survival stress is over 2.25 times greater than that of 2618. It is important to note that this material is isotropic, so the designer need not be concerned with different longitudinal and lateral strength and fatigue performance.

Table One: Comparison of Key Properties

Compared to 2618, the coefficient of thermal expansion (CTE) is 32% less and the thermal conductivity is about 7% greater. These properties make it clear why an MMC piston can be designed either for the same life with a much-reduced weight, for much longer survival at the same weight, or some combination of greater life with reduced weight.Several years ago, one CART team began using cylinder liners from 225xe to address a longevity problem in that highly-turbocharged application. The new liners lasted so well that they could be re-used during two rebuild cycles. Currently, a major piston manufacturer has developed a big-bore kit for motorcycle engines using a 225xe liner.One company offers a line of 225xe con rods for popular 250, 450 and 650cc single-cylinder 4-stroke motorcycle racing applications. These one-piece rods, for built-up crankshafts with roller bearings, originally came with a pressed-in steel outer race for the rollers. Next, they eliminated the pressed-in steel race and ran the rollers directly on the con rod MMC surface. The current products now run with NO rollers. The MMC big-end surface forms a hydrodynamic bearing against the journal, without any coatings.While no steady-state power gains have been attributable to the use of these conrods, there has been a dramatic improvement in transient acceleration performance. They are nearly half the weight of the steel pieces they replace, which allows the counterweighting on the crankshaft to be lightened dramatically in the rebalancing, significantly reducing the mass moment of inertia of the crankshaft. That reduction also serves to reduce the gyroscopic moment which the engine generates during yaw and roll manoeuvring. Some teams report being able to eliminate the counterbalance shaft which some of the engines use to reduce engine vibration. Drivers report that the engines-sans-balance shaft are no worse than the original setup.