The con rod fastener is by some margin the most highly stressed fastener in the engine, and is almost certainly exceeded in absolute stress levels only by the valve springs. In restraining the mass of the rod and piston assembly at top dead centre on the exhaust stroke, the inertia forces are responsible for the peaks on the cyclic load curve for the con rod bolted joint. How this load is translated into a stress which the rod fastener sees is a function of the design of the rod itself, the design of the bolt and its initial tightening conditions.

The worst scenario is when the initial tightening of the bolt is insufficient to keep the con rod joint loaded during maximum inertia conditions. If the two parts of the rod become separated, leaving only the fasteners to maintain them as an assembly, then the fasteners take all of the inertia load, and they are not designed to do this. Engine failure is guaranteed in this case – it is simply a matter of time.

When the fastener is correctly tightened, the joint will remain loaded, and the extra load experienced by the fastener is much smaller than the earlier scenario. The proportion of the total load to which the fastener is subjected is defined by the internal load coefficient of the joint, and this is a function of the stiffness of the fastener and that of the clamped members.

It is of paramount importance that the fastener is correctly tightened and, in the case of con rod fasteners, that this is very accurately checked by measuring the length of the fastener before and after tightening. The elongation measured in tightening is proportional to the load in the bolt, and this is a very much more accurate way to confirm the load in the bolt compared to other methods such as torque tightening or torque and angle.

For this reason there are features on the end of the rod bolts that make measuring the elongation easier when using a specially adapted micrometer, which is provided with ball-shaped ends instead of the traditional flat measuring tips. These engage with centre drillings in each end of the rod bolt, providing a far more repeatable assessment of length than simply measuring over the ends of the fastener with a standard micrometer. The improvement is something you can assess for yourself if you have both types of micrometer available.

Written by Wayne Ward

Another four-cylinder motorcycle engine, although not a race engine, also had an unusual number of con rods to control its pistons. The Wooler had a single-throw crankshaft, a single primary con rod and four secondary con rods, although the primary rod was not actually connected directly to the pistons. Whether we consider this engine to have one con rod or five, it is certainly a novel concept. Fig. 1 shows an actual Wooler engine while Fig. 2 is a cutaway drawing of it.

The primary con rod, connected to the crankshaft, actuates a lever, causing the actuation point to move back and forth along an arc. The lever is connected to a further lever to which four secondary rods are attached, which move the pistons back and forth in the bores. The pistons are arranged in pairs above each other, with two opposing banks of cylinders. The engine was arranged with the cylinders on each bank above each other and these cylinders above the crankshaft,  making the engine pretty tall. The images here show the layout of the engine, which is difficult to describe with words alone. The lubrication of the cranktrain would be problematic, as only the crankshaft end of the primary rod is in constant motion – the remaining pivot points would all experience intermittent motion.

The Wooler’s cranktrain is a complex mechanism, and is perhaps not a pretty packaging concept. The connection between crankshaft and pistons is not very stiff, and there are more moving components than are strictly necessary. All of this means we are unlikely to see anything like the 1950s Wooler 500 cc motorcycle return to the roads, and we can be almost certain that we won’t see its like on a racetrack.

However, as an unusual mechanism and use of con rods, it has been worthy of this brief examination. It seems to be a complex solution to a problem that didn't exist: there were a number of four-cylinder motorcycle engines of the same capacity which worked perfectly well with more conventional layouts although, at the time, single- and two-cylinder motorcycles were definitely the norm for road bikes as well as many racing bikes.

Fig. 1 - The Wooler engine, showing the arrangement of the primary and secondary con rods

Fig. 2 - Cutaway drawing of the engine, revealing more details of the packaging of the other components 

Written by Wayne Ward

It was a three-cylinder two-stroke diesel engine with a single crankshaft, six pistons and 12 con rods, plus six rockers that formed part of the cranktrain. Two-stroke diesels are an interesting concept; we don’t see them widely used but they are among the most efficient internal combustion engines. Even the latest energy recovery systems used in Formula One – which have been nothing short of a revolution in fuel efficiency in motorsport – cannot easily compete with a well developed two-stroke diesel. Brake thermal efficiencies in excess of 45% are possible for power units developed for commercial transport, and huge stationary engines can achieve considerably higher figures than this.

In the TS3 engine, each pair of pistons moved towards each other in a cylinder and, compared to the similar concept of opposed pistons used on the Napier Deltic (which used three crankshafts to do the same thing in a much larger package size) the Commer was a very tidy and compact design in terms of packaging. The link at the end of the article shows the engine in section, and if you refer to this or another sectioned drawing of the engine, it will make the process of understanding this unique engine much easier.

The inlet charge for each cylinder was compressed by a pair of pistons. Each piston had a short con rod attached to a very substantial rocker. In turn, the rocker was connected to the crankshaft by a further con rod. The rocker appears not to have had a 1:1 ratio – that is, the piston stroke was larger than the crankshaft stroke. The rocker also had some offset, which was necessary as the pistons in each cylinder clearly needed to be coaxial, but the motion was imparted by separate throws on the crankshaft spaced 180° apart.

So, in the cranktrain, we have a much more complex arrangement than in a normal engine, but this is offset in an opposed piston engine design by the fact that the engine has no cylinder head.

Does this have any relevance to race engine designers? Some years ago, when the possibility of hybrid systems was first being considered in Formula One, and when I was working on Formula One engine development, I took part in a discussion of what we might do if given a set amount of energy to start a race with, and no other restrictions. A turbocharged two-stroke diesel engine was mentioned at the time. If fitted with systems for energy recovery and running at near-constant speed, powering a generator, it might prove to be a  good solution. In terms of packaging, an opposed piston design such as the Commer Double Knocker might well be a good solution.

At the moment though it seems unlikely that this sort of power unit would be used in any kind of motorsport, but for passenger and freight applications a similar concept could prove to be a better solution to the ‘range extender’ engines now being developed for electric vehicles.

For sectioned views of the engine go to: https://www.flickr.com/photos/jeffspiccies/5384257681/

Written by Wayne Ward

If we deal with the first of these, this is solved relatively easily. If we interpose a bearing block between the crankshaft and the rod then we can reduce contact pressures; the block is ‘racetrack shaped’ – that is, a round-ended block – with a cylindrical bore. Both of these contacts are conformal, and the contact between the crankpin and bearing block bore is much as it would be in a conventional con rod, while the contact on the outside of the block is simply two flat surfaces in sliding contact.

It would be relatively easy to assure ourselves via calculation that our bearings can generate a proper oil film. In the case of the block running in the slot, we need to take account of the fact that the bearing block is reciprocating in the slot and comes to a stop when the crankpin is at ±90° relative to top dead centre. Coming to a stop means there is no velocity to generate an oil film, and this could lead to wear. However, with a suitably arranged feed of pressurised oil, it should be possible to preclude any wear problems.

The stiffness of the ‘big end’ of the rod is also a concern, as is its strength. In order to make the big end of the rod stiff enough, a lot of detailed design and analysis would be required, especially if the design was of the type that is split through the big end, as is the case with most four-stroke engines. The bolts would need to be large, as the bending moment would be considerable.

When the inertia loads are at their maximum, the distance from the bolts to the line of action of the applied load is much greater than in a conventional rod. However, this is mitigated to an extent by the fact that there are no secondary inertia forces, owing to the fact that the motion of the piston is simple harmonic motion. Also, because there are no concerns over high secondary forces or high side thrust loads – both of which are associated with short articulating con rods – the Scotch Yoke rod can be made as short as is practical, thereby minimising the reciprocating mass and the inertia forces associated with it.

The concept lends itself best to 180° vee engine architectures, because on such flat engines two pistons and the rod assembly operate on a single axis – that is, there is no bank offset – and some of the mass penalty of a Scotch Yoke mechanism is shared with another cylinder. However, the flat engine is not widely used in motorsport where bespoke engines are concerned. Granted, there are a great many Porsches competing in all kinds of racing, and Subarus have long been a favourite for rallying, but the unique packaging challenges of a flat engine means that the most advantageous engine configuration for the Scotch Yoke would present its own disadvantages in most race vehicles.

Written by Wayne Ward

However, the con rod poses some problems for the engine designer, especially the race engine designer who is under pressure to keep component mass, overall engine mass and the dimensions of the engine to a minimum. To satisfy all these criteria, the logical strategy is to reduce the con rod length to a absolute minimum, but on many race engines you will find that this strategy has not been followed.

There are two good reasons for this – vibration and friction, and short rods are bad on both counts. They lead to high second-order forces, which act at twice the frequency of engine rotation, and because they act at twice engine speed, they require balance shafts or similarly complex mechanisms to counteract these forces. Race engines don’t tend to bother with such power-sapping devices for increasing engine refinement, but we do need to be mindful of the level of the vibration and the effect it has on the engine, car and driver.

In terms of friction, short rods mean that the angle through which the rod has articulated when the pressure in the combustion chamber is at a maximum is increased. If we say maximum pressure is at 10° after top dead centre, an infinitely long rod will have moved only a very small angle, and the sideways thrust on the piston owing to rod angularity is correspondingly small. Very short rods have large articulation angles though, and therefore the piston skirt is thrust into the cylinder liner with much higher force, and frictional losses are increased as a result.

Engineers tend to look at the ratio of crankshaft throw to rod length. In a race engine, a ratio of 0.33 would be considered high and 0.2 at the low end of the scale. However, the choice of this ratio is chosen according to a number of factors, of which engine operating speed is the most significant.

There is a mechanism though that gives no rod articulation, very little piston side thrust and no second-order vibration. The Scotch Yoke is a simple device, and the crank is connected to the piston essentially by a straight, stiff rod without any articulating pivots. The link can be made as short as is practicable without fear of increasing friction or vibration, allowing the engine designer to make the engine as compact as possible.

In the simplest case, the crankpin simply oscillates within a slot, but this is only suitable for very low speed machines where operating forces are low. In an engine, the concept requires more thought and engineering to make it work properly. In the next article we will look at the Scotch Yoke in more detail.

Fig. 1 - A simple Scotch Yoke mechanism. The piston and slider are a single part, and the crankpin is constrained to move within a track. We can see some obvious mechanical limitations here, but evolutions of this mechanism have run successfully in engines

Written by Wayne Ward

The con rods used in Formula One are beautifully engineered and, in common with a great many racing rods, are machined all over – that is, every external surface is machined. You might imagine that this gives the mechanical engineer the maximum possible freedom in design, but there are limits as to what can be done with machining, and the rules that prohibited welding of rods caused at least one Formula One engine manufacturer to abandon well-advanced plans to race hollow con rods.

Hollow structures are often very attractive to designers as they combine stiffness and low mass. Metallic rapid prototyping methods such as direct metal laser sintering (DMLS) allow us to not only build simple hollow structures but also very complex hollow components with load-carrying internal lattice structures and small curved holes carrying coolant/lubricants.

The production processes, of which DMLS is only one, generally use finely divided powder with carefully controlled size and form, and fuses it together to form the component. In terms of a racing con rod, the main concern for the design engineer is that it is possible to remove loose powder from within the finished hollow structure.

The range of materials already commercially available suit our needs well, having been developed initially for aerospace customers, and for con rods there are various high-strength steels and titanium.

A real bonus when developing a novel, hollow con rod using rapid prototyping would be that development would be relatively fast. We can make a change to the internal structure without the need to consider new tooling, meaning that a batch consisting entirely of slightly differing and unique designs would be no more costly to ‘build’ than a batch of identical parts.

We cannot view such rapid prototyping processes as an engineer’s Utopia though; it is very likely that we would still prefer to machine many, if not all, of the external surfaces. The various rapid prototyping techniques, although suitable for making complex hollow components, are relatively slow. It lends itself therefore to small-volume production, but even there the costs would at the moment be prohibitive to all apart from those with exceptional budgets.

Motorsport has yet to go through the process of learning how best to use rapid prototyping, and finding suppliers who can make parts to the highest standard on a repeatable basis. The standard required embodies not only the obvious aims of consistent geometric accuracy and a fast, reliable service, but the less visible target of being able to achieve good material properties, particularly fatigue resistance. The initial condition of supply of the powder and its ongoing handling and condition are important aspects of making the techniques viable.

There is little doubt that we will see con rods made using such techniques in the near future, but it is also very likely that it will be some time before the speed of the production process and the associated costs are low enough to see it widely adopted for con rods.

Written by Wayne Ward

Where the rod is of the usual split type, it is generally the case that the rod and its cap will be made from the same piece of material, quite often in the form of a forging. A wide range of metallic materials are used, from aluminium through titanium to steel and even metal matrix composites. Where billet con rods are made, it is still conventional to produce the rod and cap from the same billet. However, there might be occasions where the designer has to consider a cap made not only from a different piece of material but from a different material altogether.

The reasons for doing this would have to be compelling as there could be problems with taking this approach. Where a different material is specified for rod and cap, it is will generally in a search for increased cap stiffness. If the cap is felt to lack stiffness, it might be the case that other components limit the scope to increase the stiffness of the cap through geometry changes.

For instance, in an inline four-cylinder engine, the distance between the crankpin and the closest components in the sump/lower crankcase might not allow any appreciable increase in cap stiffness. To increase the stiffness through geometry changes, we will generally look to increase the height of the ribs on the cap. Increasing the height of the ribs is the most efficient way to improve stiffness in terms of adding the least weight per unit stiffness increase.

However, where we have reached the limit of increased rib height, a material substitution for the cap might be our only option. This is really only sensible for rods made from materials that are less stiff than steel. For a steel rod, there are only a few materials that offer an appreciable increase in stiffness, and these are generally much more dense.

Compared to titanium, steel offers an increase in stiffness of 75% or more, depending on the precise grades of titanium and steel in question, but we can see that large changes in stiffness are made possible by material substitution. If the change required is less than this 75% then the mass penalty of the steel cap can be lessened by reducing the rib height.

The problems associated with using a steel cap on a titanium rod will come from the physical properties of the material. As the rod reaches working temperature, the thermal expansion of the rod relative to the cap will tend to superimpose a distorted shape on the big-end bore, which is additional to whatever dynamic distortion is present. The problem would be worse if we were to use a steel or titanium cap on an aluminium rod, owing to the greater difference in the coefficient of thermal expansion.

Written by Wayne Ward

The first element we need to fix is size, which for many engine development projects will be fixed by the size of the existing crankshaft and the availability of bearings to suit. Where a bespoke engine is concerned, we have a clean sheet. The big-end sizing will be affected partly by the con rod considerations of bearing loading, but may also be influenced by the torsional and bending behaviour of the crankshaft. If the crankpin is made too small then the crankshaft will be too flexible, and problems such as critical torsional resonances in the running range of the engine and edge loading of bearings can result.

If we assume that such concerns have been dealt with then the diameter and width of the bearing needs to be based on the calculated loads and the PV rating of the bearing material. PV is the product of pressure and velocity, and some bearing suppliers will give this data along with some advice on what limits are placed on it – a certain PV rating applies provided that either pressure, velocity or temperature don't exceed a given value. Other companies are less helpful, and either don’t have any data or won’t give you the information. Sometimes this isn’t a problem, as they want to know your operating parameters and will make a recommendation of the correct grade of bearing, providing you can give them a diameter. This can be an iterative process, as you will possibly not have the diameter fixed and want to choose the diameter and width based on bearing performance.

After you have fixed the bearing diameter and width, the next design task is to size the fasteners; again, this is done using the calculated loads. There is a wide range of thread sizes (metric and imperial) and bolt lengths to accommodate most needs. There is also a wide range of materials to choose from, starting with high-strength steels, tool steels and superalloys. You may want a bespoke bolt or stud for your application, and most con rod suppliers will be able to work with you to use a bolt you have designed and made, or will have one made for you. Sometimes, unusual thread sizes are specified – for example, for an engine where M6 is too small and M7 is too large, M6.5 bolts and suitable taps can be made. Changes to the bolt pitch and thread form can also be specified, if you have the budget for this level of optimisation.

The method by which the two parts of the con rod are locked together is the final main detail to consider. Ring dowels which are concentric with the bolt are a nice solution, but push the bolts further away from the big-end axis than other locating features. This increases the contribution of bending stresses to the overall stress in the fastener. Small-diameter steel dowel pins are often used to locate the cap to the rod, and needle bearing rollers are sometimes used for this purpose. Serrations have been discussed before and are an increasingly popular choice, with more rod makers now having the facility to cut these accurately.

 Fig. 1 - From forging to finished rod, there are lots of design choices to be made

Written by Wayne Ward

Fibre-reinforced composites are another matter entirely though. The raw material can be excellent, but the people laying it up can ruin it in a number of ways, from accidentally introducing contamination (eyelashes, arm hairs, crumbs from sandwiches) to basic mistakes in lay-up (wrong material, wrong fibre orientation and so on).

For many reasons, even if well manufactured, composites tend to have a wider variation in strength than a metal, so are less predictable. Composite properties have larger standard deviations as a percentage of the property in question. For example (not based on actual figures), if you take 100 identical samples and want the weakest to break at 800 MPa tensile stress, you might need to select a material with an average tensile strength of 900 MPa. To achieve the same result for a composite though, the required average strength might be 1100 MPa.

For these reasons, as well as those of increased manufacturing complexity, we should applaud those people who seek to make complex components such as con rods from composites. From the 1980s Polimotor project to abortive Formula One composite rod projects which actually ran in engines, the potential gains of composites mean that people will keep trying. The technology with the highest potential is continuous reinforcement, whether using a polymer or metal matrix. Both techniques have been studied for con rods.

The study by Sala (1) went as far as manufacturing trials and mechanical testing, but did not extend to putting the components into a running engine. The manufacturing technique used was squeeze-casting of aluminium around a pre-form made of graphite fibres coated with silicon-carbide. The rods were of the non-split type and so would be suitable to use with an assembled crankshaft, and were of an unusual design owing to the requirement to have the fibres running in straight lines. Sala pointed out that the manufacturing technique was very expensive owing to its complexity and manual input, although the costs of the manufacturing machinery compared very well with those for more conventional techniques such as forging or casting.

The paper by Gunyaev et al (2) took a different route, using a polymer matrix and manufacturing but with split and non-split variants. They credit previous work on the Polimotor engine and some pioneering work by Mercedes-Benz on split con rods with continuous fibre reinforcement. The split type made by Gunyaev, as might be deployed in a multi-cylinder engine with a single-piece crankshaft, was made from multiple layers of conventional multi-directional pre-preg carbon fibre reinforced polymer (CFRP). The non-split type used pre-stressing to ensure that the fibres were lightly stressed in tension at all times, and the resulting stress of the polymer matrix in the con rod in its free state was one of compression. Both types of rod were tested in engines, with varying degrees of success.

So, might we see composite con rods in engines in future? It seems so, although given the strict materials regulations in force in motorsport, it is likely to come from either a car manufacturer or a university research department. 

References

1. Sala, G., “Technology-driven Design of MMC Squeeze Cast Connecting Rods”, Journal of Science and Technology of Advanced Materials, Elsevier/IOP, 2003
2. Gunyaev, G.M., Borovskaya, S.M., and Panin, V.I., “Using CFRP for the design of a connecting rod for automobile applications”, Proc. IMechE, Journal of Automobile Engineering 1994

Written by Wayne Ward

The use of production machinery has not, however, curbed the competitive streak among racers and engineers. Production engines can benefit in terms of performance and reliability from the careful fettling of various parts of the engine – ports, for example, have long been a favourite among tuners of production engines at all levels.

Con rods are also often a target of the fettler’s tools. Steel rods are forged in huge quantities, and their surface finish and appearance can leave much to be desired. Moreover, some tuners like to ‘match’ the weight of each rod and the ‘end weights’ of the rods with a view to minimising vibrations.

In terms of appearance, forging ‘flash’ is more of an eyesore than of any real engineering concern, but it does stand out as making the rod look rough. It is common for people to want to remove this, sometimes because they labour under the impression that there are large frictional gains to be had from making the con rods smooth and shiny. The only real and quantifiable gain from improving the surface finish of the rod though is that the fatigue limit of a material improves with improving surface finish – provided that there is no residual stress in the surface of the rod. If a forged rod has also been shot-peened then any improvement in surface finish may be overcome by the reduction in fatigue limit due to the removal of the residual compressive stress due to peening.

Another point that can be missed when fettling parts is the importance of aligning any machining/filing marks with the stress field in the rod. The stresses in the beam of the rod are very much axially aligned, because the rod is loaded alternately in tension and compression. This is the direction that any remaining marks should go, if any remain after fettling. If the machining marks run perpendicular to the stress, they will act as a strong stress concentration and can lead to early failure of the rod – in fact there only needs to be one going in the wrong direction to cause a catastrophe. I have seen this exact failure on a fettled production rod.

One process that will always be worth considering after any fettling of rods is to have them shot-peened in order to put the surface into compression. The improvement in fatigue life makes the process worthwhile – anything that prevents an unwanted engine strip and rebuild following a con rod failure has to be worthwhile. Although the surface finish after shot-peening can be rougher than before the process (especially if the pre-peening finish was polished), the improvement in fatigue life from the peening more than overcomes any deterioration in surface roughness.

Rod failure can often wreck the whole engine, rendering almost all of the important parts of the engine scrap, so the decision not to peen after fettling needs to be considered carefully.

Written by Wayne Ward

The bolt hole itself is very important, not only for the reliability of the rod; it can also have an effect on that of the bolt. If we begin at the part of the hole at the end close to the head of the fastener, we obviously need to leave sufficient clearance for the bolt head. However, we need to consider carefully both the surface finish and the corner radius here, and also whether we simply plunge in with a cylindrical end mill with a corner radius or do something more ‘fancy’ in order to reduce stress concentration. Having access to finite element stress analysis is useful in such circumstances.

From the bolt head recess, we need to be sure that the chamfer provided at the top of the bolt hole is sufficient to clear the underhead radius on the bolt in all circumstances. The main part of the hole is pretty plain; we simply need to ensure that the surface finish is acceptable and that it is as perpendicular as possible to the split face of the rod.

In the other part of the hole, it is common not to have the thread extending all the way to the split line. One reason for doing this is to prevent the thread pulling at the split line. Anything that prevents a thin edge of material being present will help to avoid the possibility of a fatigue crack starting; by ‘burying’ the high stress associated with the first engaged thread, we increase the durability of the rod. Having a greater distance between the bolt head and the first engaged thread allows the shank to be a little longer and thus less stiff. This can improve the durability of the bolt. Where the bolt enters the thread, the geometry of the hole entry can affect the stress concentration factor.

Chamfers have been shown to improve the distribution of load in fastener threads by transferring some of the load from the first thread to the next few less highly loaded threads. The thread form itself is very important. There are thread forms with increased root and crest radii for both male and female parts; this can improve durability by decreasing the stress concentration at the major diameter of the female thread.

Beyond the thread engagement, the run-out into a plain counterbore at the end of the bolt is an important feature. To run the thread all the way until it breaks out onto the surface of the rod is likely to lead to fatigue failures in highly stressed rods, so a counterbore is provided which should have a radius at its base.

Written by Wayne Ward

However, no amount of hard work with a calculator or spreadsheet can hope to achieve what finite element analysis (FEA) can do for us in the optimisation of a stressed component. Even where we are able to extract an answer from a basic calculation, the ability of FEA to present the results in a graphical form and to generate graphs from the data makes its use very compelling. FEA is especially useful in the case of con rods, and a number of rod manufacturers use the software to good effect. The following is a simple overview of how such software is applied.

In terms of the simple stresses on the component, FEA can easily calculate the stress at any point and is capable of calculating fatigue data given the correct operating conditions over the whole cycle. The cyclic loads can be superposed over other loads such as the stresses due to small-end bush insertion, hydrodynamic loads and so on.

The stress and deformation around the small-end bush and big-end bearing can be monitored to prevent the designer going too far in his exertions to reduce component mass. If either bearing suffers from insufficient support, it will deflect easily under pressure and lead to excessive pressure on better-supported areas, causing failure. Conversely, FEA can prove to be a useful tool in determining where mass can be removed without affecting bearing support.

The deformation of the rod can be calculated so that one includes the correct figure in the calculations for valve-to-piston clearance. We can make a stab at this with hand calculations but this is either laborious if we try to achieve some semblance of accuracy, or we make assumptions based on an average beam section – which will definitely be incorrect.

If the mass and the moment of inertia of the piston assembly components are known then FEA can calculate the natural frequencies of the rod in bending and torsion. Some rod makers who use FEA routinely will ask for this information (and should do if you have asked them to calculate the natural frequencies).

FEA can give us a lot of information about fastener loads and stresses, including predictions of joint separation. A promising new high-speed engine concept that I was involved with came to grief when we realised through the use of FEA that we wouldn’t be able to keep the rod joint together at the proposed engine speeds.

Written by Wayne Ward

Con rod materials have therefore remained as they were in 2007, unless someone has convinced the FIA that their material is at the root of reliability woes. The new engine formula in 2014 however offers the possibility to use something different. There is no telling what materials will be used; all we know of the engines are a few quotes from the manufacturers and some computer images of the units themselves.

Titanium has been widely used for motorsport rods for many years, and talking to a rod expert in the past few years, his feeling has been that titanium still offers the optimum rod – for a given application, the titanium rod would be lightest. Although there are many titanium alloys which could be considered, the ‘workhorse’ Ti-6Al-4V has for many years been the material of choice.

Steels are perhaps not a particularly glamorous choice for motorsport con rods, but they do offer several advantages over titanium, which means that the mass differential between the two is not as great as we might imagine from looking at some basic properties. Whatever steels would be chosen for a Formula One rod, it would probably not be one from which commercially available steel rods are made. Although there are cost constraints in Formula One, the teams can afford to spend more on their materials and components than the average motorsport engine suppliers.

On the face of it, these might be the limits of our material choices. The provisional 2014 rules state: “Connecting rods must be manufactured from iron or titanium-based alloys”; the rules further define what this means: “X Based Alloy [for example, Ni-based alloy] – X must be the most abundant element in the alloy on a %w/w basis. The minimum possible weight percent of the element X must always be greater than the maximum possible of each of the other individual elements present in the alloy.”

So, if we think about it, this offers some significant scope for development. The definitions of other types of materials would allow a titanium or iron alloy to contain a significant proportion, but less than 50%, by volume of intermetallics. Such intermetallics, for example titanium aluminide or iron aluminide, can offer significant advantages over conventional alloys. Before it was banned, titanium aluminide found favour in Formula One for valves, as it was stiffer and less dense than the titanium it replaced. It has been used as a con rod material in motorsport, although not widely. A proportion of 49% by volume of titanium aluminide in a titanium alloy would be legal under the FIA definition, and would be much stiffer than a conventional Ti alloy. High-aluminide content titanium alloys were developed with Formula One in mind some years ago, and are commercially available, though not in large quantities.

Iron aluminide was the subject of a lot of research in the 1980s but fell from favour after companies realised it was not going to be able to replace expensive superalloys. However, it remains an attractive material and there are development projects with the aim of producing high-strength, high-stiffness, ductile iron aluminide materials industrially.

Written by Wayne Ward

If we decided to look beyond traditional steels for rod manufacture and towards some of the more exotic steels available, could we make something to rival titanium? Without going into a major design study, we can’t know for certain, but we should not write steel off.

One advantage of using steel rather than titanium is that we can make both ends of the rod physically smaller while maintaining the required stiffness. At the small end of the rod, this may mean that we can push the piston pin axis closer to the piston crown. There are two possible advantages here. We may either run a longer rod with a less extreme ratio of crankshaft throw to con rod length, which should reduce piston thrust load through lower maximum rod articulation angle. The second is that we can maintain the same rod length and reduce the height of the block. Of course, this is often impractical unless we are at the early design stage of the engine project.

At the small end of the rod, it is common to run a steel rod without an interference-fit little-end bush, which allows the small end to be made even smaller. Titanium requires a bush to be used, and the section around the small end has to cope with the interference stresses and an allowance for possible damage due to galling as the bush is fitted.

The stiffer material means the load coefficient of the big-end fasteners is lower, so the fasteners are less stressed. We could therefore use smaller fasteners, positioning their axes closer to the big-end axis and reducing bending loads. When designing a vee engine from scratch, for example, we might be able to lower crankshaft height.

Given the advances in CNC machining in recent years and the opportunities to fully machine all surfaces of the rod, we should be able to produce the main beam of the rod with enough axial and torsional stiffness to rival a titanium rod without committing excess mass to the design. Steels in general have an elastic modulus of about 210 GPa and a density of around 7.85g/cc, giving a ‘specific modulus’ of 26.75; by contrast, titanium alloy Ti-6Al-4V has a specific modulus of 25.73. So, in order to produce a beam of a certain length and with a desired axial stiffness, the steel part would be lighter.

Remember though that these figures are for a general engineering steel. We don’t have to look too hard to find a steel with a significantly higher specific modulus which, for all the design areas discussed above, will offer an even greater advantage compared to titanium.

An optimised steel rod will require expensive material, more machining time and quite possibly various heat treatments and surface treatments, so it is not likely to be cheap – in fact it may be far more expensive than a titanium rod. 

Written by Wayne Ward

The most common configuration of split con rods is to have the bolts inserted from the ‘bottom’ of the rod, with the bolt head pointing away from the small end. This can be a fundamental disadvantage for some bespoke race engines.

When we design a race engine, we want maximum performance from a small, light machine. It did not take man very long to realise that, in this respect, a vee engine is a good idea. Many purpose-designed race engines are vees, as are a number of engines in performance passenger cars, especially naturally aspirated ones. When we take a split con rod and plot its locus, when used in a vee engine it is quite often the head of the con rod bolt or nut that defines the lowest part of the locus volume. If we design the sump to maintain the smallest possible clearance to the rod locus, we can see that the bolt head defines the position of the crankshaft axis relative to the bottom of the engine.

In trying to minimise the centre-of-gravity height of the engine, the crankshaft should be as close to the bottom of the engine as possible. In a single-seater chassis, the engine is mounted as low as possible, so the rod locus affects how low the engine can be installed in the chassis. The FIA imposed a minimum dimension for Formula One engines some years ago for the crankshaft height, and also imposed a limit on centre-of-gravity height in order to prevent costly development of ever-smaller engines.

Also sometimes a concern in very compact vee engines is the fact that the rod bolt on the rod attached to a piston on the left-hand bank of the engine can come close to the cylinder liner on the opposite bank of the engine (and vice versa).

Both of these problems can be solved in part by using rods with ‘inverted’ bolts – that is, where the fasteners are inserted so that the bolt heads are closest to the small end of the rod. Such rods, if well designed, have a smaller rod locus than their more conventional brethren and so allow us scope to design an engine with a smaller crankshaft axis height. A rod fitted with inverted bolts has a disadvantage though in that the bolts can be awkward to fit and tighten, but this may be a small price to pay for an improved engine installation.

Fig. 1 - A racing con rod with ‘inverted’ bolts may allow the engine designer to lower the centre-of-gravity height of the engine

Written by Wayne Ward

As with many components in a modern road engine, economics play just as important a role as the engineering. Engineering parts and assemblies down to a cost is a real skill, which goes far beyond getting a few quotes against a drawing. The economics of con rod manufacture dictates that the parts have as few machining steps as possible. Ideally, for a split (two-piece) con rod, a basic forging will have very little machining carried out, with the bores on the small-end and big-end axes machined to size, thrust faces skimmed to width, the big-end fracture split and the fastener holes finished (though not necessarily in that order).

Conversely, a racing rod is often machined all over. We can expect this to be the case with parts produced from billet. However, it is quite common for forged con rod blanks to be turned into racing rods without a single square millimetre of the original forged surface remaining untouched by a machining cutter. There are two main reasons for this.

The first reason is reliability. Parts with machined surface finishes perform better in terms of fatigue strength than those with cast or forged surfaces. Fatigue crack initiation takes place at lower levels of applied stress (including any residual stresses) on rougher surfaces. Many engineering textbooks which cover basic fatigue calculations will include surface finish factors. For two parts of the same material and undergoing identical stress cycles, a part with a forged finish will fail before one with a machined finish.

The second reason is consistency of manufacture and its effect on component mass. A batch of parts which have been accurately machined will have a much smaller spread of wall thickness than a part whose dimensions are controlled by one or two forged surfaces. We can relatively comfortably specify a rod width of ±0.05 mm for a machined part, safe in the knowledge that this is quite easy to achieve. If we need to rely on ±0.25 mm for a forged part, we need to leave a further factor of safety against fatigue, because our service load must be borne by a potentially smaller area. There is also an equal chance that, having calculated stresses based on the smallest possible area, the parts are on the large side. The amount by which a forged con rod must be over-engineered to cope with manufacturing tolerances is much greater than is the case for a machined rod.

For these reasons, it is likely that fully machined racing rods will remain common, and for reasons of economics we will continue to see rods with very little machining in road vehicles.

Written by Wayne Ward

However, in the roadcar industry, where costs are more important, it is very common to use a different method for steel rods - fracture-splitting. The idea is simple: a defect is deliberately introduced to the rod and, with the big end supported close to the split line, the rod is struck laterally. The high strain rate causes an otherwise suitably ductile material to behave in a brittle manner. The fracture produced shows no deformation, and the surface shows a random rough texture. This is very far from what we produce with the usual precision machining used in the production of racing rods.

The technique has significant merits compared to precision machining. The significant cost reduction is clearly attractive, especially during a period when everyone is looking to cut costs. However, lack of control over the contour of the surface does not in this case mean lack of precision. A fracture-split con rod has two parts that mate absolutely perfectly together. Also, in the same way that no two snowflakes or fingerprints are the same, there are no two identical fracture surfaces. For an experienced engine builder, it should be impossible to assemble a rod and cap that couldn't be assembled without it being obvious that something was wrong.

Given these desirable traits, why are fracture-split rods not in common use in motorsport? It isn't a method that can be applied to every material; many materials do not exhibit sufficient strain-rate sensitivity to allow them to fracture every time. One of the main 'crackable' steel materials in use for production engines is C70, a high-carbon wrought steel. The high-carbon composition reduces the content of ductile phases in the steel. This steel can be fractured using special machinery at room temperature.

Powder metallurgy materials have also been developed especially for fracture-split con rods. The powder metal materials developed for rods are favoured for larger, low-revving engines as produced for the US market, while European and Asian engines tend to be smaller and higher-revving, and these tend to use the wrought materials. However, the fatigue strengths of these materials are not as high as we would expect from high-quality wrought materials used for racing rods. For high-performance motorcycles, a method has been developed for fracture-splitting carburised con rods. Yamaha introduced the method for the 2003 R1, and found a 30% rod cost reduction compared to the previous machined split-face rod.

Written by Wayne Ward

The bolt is often preferred, owing to lower parts count and lower cost. The head of a bolt is often physically smaller in terms of diameter and depth, although using a bolt with a very shallow head height can bring its own problems in terms of increased stress concentration and lack of resistance to head-rounding during tightening or disassembly. Providing that the bolt is designed with sensible proportions, one might question why anyone might want to use the stud and nut combination.

However, there are valid reasons why some engineers might choose to use a stud and nut rather than a bolt. If the engine is expected to be rebuilt, and you need to re-use con rods, you may want to minimise damage to the internal threads in the rod by leaving the male fastener installed permanently. This is a consideration for titanium con rods, whose threads are more easily damaged than their steel counterparts. The stud-and-nut combination also offers the opportunity to dispense entirely with a female thread in the con rod, thereby eliminating any possible thread damage and also removing the stress concentration associated with it. In such cases, the stud has to be designed with an anti-rotation feature, and this is often achieved by using a D-shaped head.

A further reason for the use of a stud and nut is that the stress field in the fastener differs from that in a bolt. A nut and stud can be more flexible than a bolt, especially if a nut material of relatively low modulus is chosen. This can put less bending load into the fastener. The flexibility in terms of nut geometry and modulus means the stress concentration at the first thread within the nut can be minimised. Where a male fastener is used with a nut of lower modulus, the stress concentration is reduced as the female threads flex more, thus bringing more of the female threads into use and improving the load distribution over the length of the nut.

Studs can also incorporate location diameters, precluding the need to provide separate dowel pins or serrations. While the same features could be added to a bolt, this is often not the case as any variability in interference can add extra friction into the bolt-tightening procedure. While con rod fasteners are most commonly pre-loaded using stretch control, increased torque requirements to overcome the friction at the interface between a rotating location diameter and the static con rod will increase shear stress during tightening.

Written by Wayne Ward

One problem with thrust bearings is the fact that, in their simplest form, they comprise a stationary disc in sliding contact with a moving disc. If the plates remain a set distance apart, they do not generate any oil film pressure and so cannot act as a bearing. It is only by the relative movement of the bearings relative to each other that squeeze-film lubrication operates, producing pressure which counteracts the movement of the plates towards each other.

The recent article on crankshaft rod thrust face design mentions that there are features that can be used on the rod thrust face to help generate pressure in the oil film. The article mentioned that a discontinuous face on the rod thrust bearing of the crankshaft would help generate oil film pressure between the crankpin thrust faces and the corresponding thrust faces on the con rod.

We can often see this when looking at racing con rods as a series of shallow and relatively narrow grooves in the thrust face of the con rod; where two grooves are used they are commonly placed over the split line of the rod. This not only adds a feature that can aid lubrication, it removes sharp edges at the split line from the thrust face. As we will see, having sharp edges is not necessarily a bad thing in this situation.

Where a fluid is present in the gap between two parallel plates moving relative to each other, and the gap between the plates suddenly narrows, there is an increase in oil pressure. This effect isn't restricted to liquids; the same effect can be observed in gases too, which can form an effective bearing between parallel flat surfaces when a step (or multiple steps) are introduced.

This type of arrangement is known as a Rayleigh step bearing, after the famous physicist Lord Rayleigh. While the arrangement seen on many con rods with two or more narrow grooves is common, it is possible to be much more aggressive if the rod thrust face arrangement is designed with the principles of the Rayleigh step bearing in mind. The pressure generated in the bearing is greatest if the edge of the step is sharp, and there are some general design rules that optimise design ratios for maximum generated pressure.

If the thrust face is designed as a series of small Rayleigh step bearings around the periphery of the rod bore, a lighter big end can be achieved. On a vee engine where two rods share one crankpin, this principle of 'aggressive lightening' by using a series of small step bearings can only be reliably applied to one side of the con rod; the thrust faces between the two rods are best kept as relatively plain, flat faces.

While not specifically designed as effective Rayleigh step bearings, grooves across the face of crank thrust bearings act in the same way.

Written by Wayne Ward

As race engine engineers, we generally take great care to keep both piston assembly mass and crankshaft inertia to a minimum. Something that might seem a very attractive concept, therefore, is to shorten the con rod. Not only would we save mass in the con rod, but the block could be made shorter, reducing the mass of the block and lowering the height of the centre of gravity of the engine. Again, these are all generally laudable aims for a race engine.

However, we find that there are a number of reasons to maintain or increase con rod length. The first of these is the mitigation of secondary forces. These secondary forces act at twice engine speed and are proportional to the ratio of crank throw to rod length. We can see that decreasing this ratio will reduce secondary forces and, with a fixed crank throw, we would look to increase rod length to achieve this.

In the formula above, F is the force due to the reciprocating mass, m is the reciprocating mass, omega is the rotational speed of the engine (radians per second), r is the crankshaft throw, l is the rod length and theta is the angle from TDC. At TDC, cosθ = cos2θ = 1

For a 100 mm rod acting on a crankshaft of 40 mm stroke (20 mm throw), the secondary forces add 20% to the primary force. Where short rods are used, the r/l ratio may be as much as 0.3.

The second reason why short rods are not often used is the effect of rod angularity. The longer the con rod, the lower its angle to the cylinder axis for any angle of crankshaft rotation. The higher the r/l ratio, the greater the angle of the rod when maximum cylinder pressure occurs.

The effect of higher rod angularity is to increase the side load on the piston skirt. For a given coefficient of friction, a higher load means a higher drag on the piston and increased frictional losses. The interaction between the skirt, lubricant film and cylinder wall means the coefficient of friction cannot be assumed to remain constant, but the principle holds true that lower con rod angularity offers lower friction.

There are other ways of achieving this, such as pistons with pin bores that are offset from the piston axis, or engines with cylinder axes offset from the crankshaft axis. Each alternative, however, comes with its own difficulties and complexity of design. A longer con rod is often compatible with all existing components, and unlike offset engine layouts, the crankshaft stroke and piston stroke remain identical.

Fig. 1 - Even where low mass is most highly prized, using the shortest rod possible isn't necessarily the preferred option

Written by Wayne Ward

However, for the most highly stressed fasteners in the engine - as in the con rod bolts or studs - most people wouldn't consider designing their own fasteners. Con rod fasteners are very highly engineered, are made from very high strength materials and there are a number of significant risks for those who design their own and have them manufactured. The three most serious problems that someone is likely to encounter in having a bespoke rod bolt are deficiencies in design, manufacturing problems and incorrect material specification and heat treatment.

Anyone tempted to design their own con rod bolts or studs should select a manufacturer who is well-versed in their manufacture. Producing consistently high-quality rolled threads on very high strength materials is not easy. There are also a range of other production processes used on con rod bolts that are not often used for other fasteners.

If we turn to specific design points, there are a number of things that are specific to con rod fasteners which we wouldn't find in a commercial bolt. There is often a location diameter close to the head of the fastener, to ensure proper alignment. Where a bolt is used, the head itself is normally a 12-point head rather than a socket head or hexagon head, as they have a greater torque capacity than a socket or hex head.

The head itself is carefully designed so that there is no sharp edge on the seating face that might dig into the con rod cap and cause a stress concentration. The shank diameter of the bolt is, for a very good reason, carefully controlled by design and manufacture in terms of diameter and length.

Most con rod fasteners have their pre-load controlled by measuring the length of the bolt or stud. By knowing the axial stiffness of the fastener and accurately measuring the stretch, we can calculate the load. As the axial stiffness of the shank section is directly proportional to the length of the shank and the square of the diameter of the shank, we need to control the shank length and diameter very carefully if loads are to be calculated with any accuracy. Also, in order to help the engine builder measure the stretch in the fastener, both ends have specific design features that allow accurate measurement with a specially adapted micrometer.

The manufacturing processes are critical to the success of the fastener. Among fastener manufacturers there is some disagreement on the need to cold-forge the head, but in producing a 12-point head, forging is necessary. Great attention must be paid to the underhead fillet, and these areas are often fillet-rolled to impart compressive residual stresses.
The surface finish of the fastener is also important. Naturally we want the surface to be free of any blemishes that might produce a fatigue crack initiation point. However, a mirror finish is not necessarily what we are looking for; we want a surface finish that produces a consistent level of friction. While we don't control con rod pre-load by torque, lower friction produces lower torsional shear stresses during tightening and operation.

Fig. 1 - If you want your rod to last, don't take risks with the fasteners (Courtesy of L Travers)

Written by Wayne Ward

Certainly the latest con rods for high-level motorsport would perhaps open Cameron's eyes a little, but he would probably still see them as art. The relationship between shape and stress has been somewhat clouded by other important considerations. Con rods are not, on the whole, highly stressed parts/assemblies. Their shape is defined by providing direct load paths. Clearly, stress is a very important consideration in certain areas of the rod, especially where assembled and bolted rods are concerned.

However, much of a con rod's design is concerned with stiffness. A con rod has more than one stiffness and many natural frequencies. How many of these natural frequencies are relevant depends on their frequency and the operating speed of the engine, which serves to excite the rod, eliciting a response. Axial stiffness is an important consideration if valve-to-piston clearances are marginal. Many people push these clearances to the limit in order to achieve high compression ratios.

Torsional stiffness is an often neglected quantity; the torsional resonance of a con rod can also have an effect on compression ratio, as it affects the clearance around the circumference of the valve to the pockets provided. Large resonant amplitudes require large pockets, and these serve to lower compression ratio. Calculating torsional stiffness and natural frequencies can be as important for a con rod as a crankshaft. If a major torsional mode is excited by an engine speed within the operating range, we had better try to avoid constantly running at this engine speed, lest con rod torsional failures ensue.

In the past, designers of crankshafts for ships were required by law to undertake extensive calculations of the torsional response of the crankshaft/propeller shaft system to excitation, and to recommend engine speeds that must be avoided except in transient conditions. If we are going to design con rods for minimum mass, we ought to think about doing the same calculations for the con rods and their associated components. Some con rod companies who are expert in such things may ask for details of piston assembly inertia to aid them with their calculations.

On the internet you can easily find pictures of Formula One con rods, which range from the quite simple to the very complex. All are designed with considerations of torsional stiffness and natural frequency in mind. However, some of the more outlandish designs, which might appear to incorporate strange shapes and design features in the beam, are conceived to show the kind of features that may be produced for reasons of beam stiffening.

Written by Wayne Ward

At the recent Autosport Engineering show in Birmingham, England, I discussed the matter of rod coatings with a supplier, who showed me a couple of con rods he had to hand. The supplier explained to me that the con rods were coated with diamond-like carbon (DLC) with the aim of reducing friction and wear of the thrust faces of the big end. Other coatings have been used over the years for the same purpose and with some success. Metallic molybdenum has been used for a long time, although this has been supplanted to a large extent by the very hard, thin and lower-friction engineering coating processes such as DLC and chromium nitride (CrN).

DLC has an advantage over CrN and the metallic coatings in that it has a very low coefficient of friction. While this should have little advantage once conditions for full hydrodynamic lubrication are satisfied, lower friction and improved wear characteristics are useful at start-up and when lubrication is in the mixed regime.

In addition to the application of DLC to big-end thrust faces, it could also be usefully applied on the thrust faces of con rods at the small end. In this application, where the motion between sliding surfaces is at relatively low velocity and where the motion is intermittent, such coatings might prove to be very useful.

There is a further use for low-friction coatings for the manufacture of con rods with interfered small-end bushes, especially in highly stressed applications that use titanium for the con rod. Titanium is prized for its combination of strength and low density, and has found common use in racing as well as an increasing acceptance as a material for production car and motorcycle con rods.

Where a bush is pressed into the small end of a titanium con rod, there is a chance that the bush may 'pick up' in the bore of the rod, with the consequence that the bore of the rod is damaged. Depending on the location of the damage, this may be of little significance, but there is always the chance that the damage is in a critical location which could cause the rod to fail. For such applications a DLC coating applied to the bore of the small end of the rod can be used to reduce the chances of such damage occurring. If we can prevent the damage happening, we don't have a 'sore spot' from which a fatigue crack could emanate.

In addition, the use of coatings in the bore of a con rod could allow the rod to run 'bushless'. This has been discussed a number of times, but the main application of this design philosophy has been with steel rods. If the same could be achieved with a titanium rod then we could reduce the size of the small end while reducing the risk of bore damage by not having to fit bushes.

Fig. 1 - DLC-coated rods spotted at a trade show recently

Written by Wayne Ward

There are a number of advantages to running a con rod without an interfered small-end bush. Obviously there is a lower parts inventory at the rod manufacturer for a start, but there are a number of good engineering reasons for looking to a bushless con rod.

There is a small weight saving gained from dispensing with the bush itself. The small end of the rod can simply be maintained in section, but brought closer to the pin axis. However, it can be further optimised as it no longer has to cope with the stresses induced by the interference fit of the bush, in addition to any service loads. The section of the small end can safely be reduced while still staying within safe working limits of stress.

Owing to the fact that the small end is now smaller, we might choose to push the pin bore higher in the piston and run a slightly longer rod. One advantage of this is that the angularity of the rod is decreased, and piston-to-bore friction can be reduced as a result. A second advantage is that secondary forces are reduced; these are a function of the crank radius-to-rod length ratio.

Alternatively, we can simply re-optimise the design of the rod in the light of having a lower-mass small end. A consequence here is that the crank counterbalance mass can also be reduced. It is unlikely that many people would re-optimise all of these components, having decided to equip their engines with bushless con rods, but for those in the business of regularly designing bespoke race engines, such detailed re-optimisation is possible and is an attractive prospect, even when overall engine weight is controlled or where component mass is controlled.

A wide range of materials have been used with success for bushless con rods, including metal matrix composites and aluminium alloys. Materials such as aluminium and titanium can suffer if an interfered bush does not go in 'cleanly'. If there is pick-up or galling when the bush is interfered, this can act as a stress raiser and therefore the factor of safety against failure needs to take some account of this possibility. Attempts to prevent such damage during fitting involve use of special lubricants, or coatings for both the outside diameter of the bush and the bore of the small end.

In addition to these reasons, the rod should be simpler, quicker and less expensive to make. There is a reduction in the number of precision operations involved and a lower parts count, both of which should result in reduced manufacturing time and costs.

Fig. 1 - Bushless con rods for race engines have a number of advantages

Written by Wayne Ward

Traditionally, racing rods have had their mating caps located by either the use of small dowel pins, or by ring dowels. To accurately and repeatably locate the two pieces of the rod assembly together, there is a requirement to use a location dowel on each side of the rod. A third option of using slightly interfering location diameters on the fasteners is only generally used where the fastener is used in combination with a nut.

In recent years, new methods of location have become more popular for bespoke rods; precisely machined serrations on the rod and cap locate together to produce a very rigid joint. The advantages are that any shuffling and fretting of the rod split face can be avoided, and there is a possibility to get rid of the location dowels, giving more freedom in the design of the rod.

There are two variations on this technique: one is to use straight-cut serrations, similar to a very short section of a rack, and the other is to use curved serrations (see accompanying pictures).

The straight-cut serration can only constrain the cap in a single direction, and so would require further features such as ring dowels or location pins to accurately locate the two pieces of the con rod. Given that the width of the con rod big end generally relies on the two parts of the rod being located precisely, the danger in not providing extra location features is that rod thrust clearance is reduced from design values. The straight-cut serration shown in Fig. 2 has a much smaller number of serrations than the curved serrations produced by another manufacturer. It should be noted that both of these types of serrated joint face are being supplied to and used in high-level motorsport.

Curved serrations constrain the cap from movement in any direction on the split plane of the rod, and so are free of the requirement to use any further location features. There are, of course, an infinite number of ways of arranging the serrations; Fig. 1 shows one solution, with the serrations for each side of the con rod being centred about an axis that is outside the body of the rod.

Providing that the central axes of the curved serrations on each side of the rod are not coincident, the cap can not move in a planar motion or rotate, thus tying down all degrees of freedom. The rod with curved serrations shown is an aluminium rod. Many of us will be familiar with the problems of running two aluminium faces in contact - fretting is a particular problem, which is likely to be mitigated by controlling the very small movement causing this phenomenon.

Figs. 1 and 2 - Two options for joint face serrations are shown. Both curved and straight-cut serrations are currently used in high-level motorsport

Written by Wayne Ward

There has been a trend in modern engines for production vehicles, in part because of the relentless drive to cut costs, to dispense with the tags/tangs on big-end bearings and, of course, with their corresponding slots. The elimination of a machining operation on both the rod and its cap reduce the cost of the con rod. The method of producing the tag in the bearing is a deformation process rather than machining, and there is a cost saving when eliminating this process.

Of course, a large part of motorsport is based on the use of production engines and parts, so some con rods with tagless bearings are already being used. Could the same cost reduction strategy for con rods work in racing?

In terms of preventing rotation of the bearings, this should be taken care of by the proper sizing of the con rod bore and the amount of preload applied to the bearings. The major risk in deleting the bearing tag and the slots in the con rod big-end bore is the loss of precise axial location of the bearing.

As we know, race engines are generally designed to be as small and compact as possible, and the stressed components are designed with component fatigue life at the forefront. The axial location of the big-end bearing can be of critical importance for a number of reasons. It is essential that the edge of the bearing shell does not encroach on the crankpin fillet radii, and any loss of precision in location means the design clearance between the bearing and the fillet must be increased.

In limited space, this means that either the crankpin fillet radii must be decreased or the width of the bearing must be decreased. Both of these have an impact on the life of the engine, either by possibly reducing the endurance limit of the crankshaft or by increasing the service pressures on the bearing.

Where the small end of the con rod is fed by pressurised oil from the big end, poor axial location could restrict the flow of oil, or possibly stop it entirely if the oil feed hole in the bearing and the rod are seriously out of alignment.

Of course, where the centres of the bearing shells are out of alignment relative to the applied load from the small end, and also to the neutral axis of the rod, bending stresses will be increased.

For these reasons, it is unlikely that we will see widespread use of tagless bearings in bespoke race engines in the near future.

Fig. 1 - Bearing tag slots, as seen here, are likely to remain a feature of racing con rods (Courtesy of Arrow Precision)

Written by Wayne Ward

Many years ago, this was indeed the way it was done, with the bearings manually 'scraped' by hand to achieve the correct size. Progress in the production of bearing shells has thankfully relieved us of the burden of manual bearing scraping, although many of us will possess a set of bearing scrapers that double as fantastic deburring tools. At that time, the bearing material was applied to much greater thicknesses and the bearings wouldn't work anything like as well as they do today. Part of the reason for the longevity and load capacity of modern shell bearings comes from the newer materials, but much also comes from the very thin coating and the precision with which these parts are made.

So, the first point to be established is whether it is possible to accurately deposit bearing coatings directly to the con rod and cap. After having discussed this with more than one company, this is definitely the case, although it is currently much more expensive than producing shells.

Does it work? Apparently yes. There is no technical reason that it shouldn't, and history also suggests that it should work fine. Racing companies have looked seriously enough at this idea in the past to have made and tested rods, at great expense.

The technical advantages are numerous - there are no features required to prevent bearing rotation, the rods can be made smaller and lighter, the bolts can be brought closer to the big-end axis and are therefore subject to lower bending stresses for the same applied service loads. During the period when there were no restrictions on crank centreline height from the base of the engine in Formula One racing, an advantage could be found by dispensing with the bearing shells and reducing this dimension to a minimum. On a 90º vee engine, the rod bolt (or nut) is generally the part of the rod locus that is the lowest part, and thus limits how close the crank axis can be to the lowest inner surface of the sump.

The reason why con rods with bearing coatings directly applied are unlikely to be seen widely in motor racing in the near future is one of economics. The accuracy required for the manufacture of the con rod bore would be increased, and the mechanical masking necessary to prevent coating being applied to areas other than the bearing bore would be expensive to produce and use. The very obvious sticking point is that when the bearing wears or shows signs of distress, the rod may well be scrap, or at least require serious re-work. To run engines with bearing coatings directly applied to the big-end bore would require a large increase in the inventory of con rods held by racing companies.

However, for production engines, it is very likely that this process will come to be the norm. With large production quantities, engines able to run for the life of the vehicle without disassembly and rebuild, and reduced parts inventory, the possibility to save money is very real indeed. The question of mechanical masking would be less important, as the costs would be amortised over a huge number of parts, and it is probable that the entire process would be automated.

Fig. 1 - In racing, we are unlikely to see the widespread use of con rods with bearing metals directly applied

Written by Wayne Ward

In providing adequate lubrication (and therefore very low friction) for the big end, we aid the con rod in its desire to move parallel to the crankshaft. We therefore need to provide a component for it to react (or thrust) against. Con rods are often described as being 'crank-guided' or 'piston-guided', and these descriptions refer to the component that provides the restraint to the con rod's ambition to move axially.

Where a rod is piston-guided, a surface on the rod reacts against a machined surface on the piston. In order to prevent wear of the piston, the surface of the con rod which is used to thrust against the piston needs to have a reasonable amount of surface area, and may be equipped with features that encourage an oil film to form. This is important given the small relative sliding velocity between surfaces at the little end and the reciprocating nature of the sliding contact.

The thrust surface on the rod can be in the form of the small-end bush being split into two 'top hat' bushes, thus being of the same material (commonly a copper alloy) as the small end bush. An engineering coating can be used on the thrust faces, and in doing so this dispenses with the need for multiple top-hat bushes. Chromium nitride (CrN) is a good candidate for this application, and while some people contend that no coating is really necessary, it might be a desirable option.

The big end, on a crank-guided rod, has similar needs in terms of reacting to a load, but with greater potential surface area and higher sliding velocities, the matter of lubrication is a much easier problem to solve. When the engineer has control of both the crankshaft and the con rod designs, there is scope to design an optimal solution here in terms of assembly mass and lubrication. Given the sliding velocity in the thrust contact, there is very little requirement in terms of thrust bearing area.

Where two crank-guided con rods sit on a single crankpin, as is common in vee engines, there is a thrust contact between the adjacent rods, and this is a different consideration to the crank contact. In common with the thrust contact of a piston-guided rod, the thrust contact between rods is one of reciprocating rotation over a relatively small angle and with low sliding velocities, and the thrust-bearing faces need to be considered in light of this fact. Consequently, rod-to-rod contact faces commonly have larger contact faces than do rod-to-crank faces.

In both circumstances, thrust-face coatings are common, although by no means universal, and the wear resistance of the substrate material is one consideration when assessing the need for a thrust-face coating. Sprayed metallic coatings (especially molybdenum) are commonly used, although there are a number of others based on different alloys or ceramics which could be used. Again chromium nitride (CrN) has found use for this application.

Fig. 1 - Con rod thrust faces are commonly coated to prevent wear

Written by Wayne Ward

If we improve the amount of energy that we convert, we will produce a greater amount of heat that we will need to manage, and consequently we will need to deal with higher component temperatures. Pistons are one of the components that 'suffer' in this way, and temperature can cause a number of problems. Where piston clearances are tight, the extra heat can cause the top land to scuff, and there will be a requirement to adjust the top land profile to cure this problem. While something of an annoyance, this problem is easily solved.

Of more pressing concern is the situation where piston temperatures become so high that we exceed the capabilities of the material from which it is made. Sometimes a material substitution can help here, but where we are perhaps already using a premium piston alloy that we aren't easily going to improve upon, we need to adopt a different strategy. A common solution to the problem of maintaining piston temperatures within acceptable boundaries is to spray the underside of the crown with oil. This is generally achieved by taking an oil feed from a gallery and providing a suitably aimed jet of oil to the area requiring cooling. However, this isn't always easily achieved.

Con rods can provide a solution here, as the big-end bearing is invariably fed with pressurised oil. It is common to direct some of this via drillings to the little end to ensure adequate lubrication, but we can also take a feed through the bearing to drillings along the flank of the rod and direct it to the areas of the piston crown that require cooling. Clearly, given the articulation of the rod relative to the crown of the piston, such jets will sweep out a line across the piston, but this is little different from the fixed jets that are generally fed from a gallery in the cylinder block.

This method has the advantage that no alterations are required to the cylinder block, where the design of the casting may not lend itself to the provision of machined areas, drillings and tapped holes. Depending on their design, it may even be possible to adapt existing con rods and bearings to provide the oil feeds required. The method has been successfully used in both high output series production engines and in racing. Where existing, normally aspirated engines have subsequently been turbocharged, this method might prove to be the easiest way to provide oil cooling to the pistons.

Fig. 1 - This con rod from the Honda NSX was remarkable for its pioneering use of titanium in roadcars. It also used through-rod piston cooling sprays, one of which is circled in this picture

Written by Wayne Ward

The merits of using multiple cylinders are well known, and to make compact engines with larger numbers of cylinders, 'vee' engines are a good solution. Most bespoke race vee engines have two con rods running on one crankpin, with each rod offset from the centre of the crankpin. The rod is positioned centrally with respect to the piston that operates it, and therefore the cylinders on each bank are offset by the same amount, as are the con rods.

Selecting the correct bearing size is dictated by the combustion pressure and the bore area; for a given pressure limit on a bearing and surface speed, bearing area is proportional to piston area. In general, we can therefore say that bigger-bore engines will require bigger bearings and, as bearing width dictates the width of the big end of the rod and therefore rod offset, the bank-to-bank stagger of the engine will be correspondingly larger. For a given engine capacity, bigger bores mean a shorter stroke, giving the opportunity to run to higher engine speed and increase power output. Bank-to-bank stagger has an effect not only on engine length, but also on engine mass. It stands to reason that, all other things being equal, a longer engine is a heavier engine.

Engine length also has an effect on engine stiffness and, where engines are a structural member in the chassis (fully or semi-stressed), any increase in engine stiffness is generally welcomed by the chassis designer.

Even where engine mass is limited to a certain minimum, as in Formula One, there is an advantage in having a shorter engine. While the potential mass saving is not realised, the 'spare' material can be used to advantage to produce a stiffer engine.

Given that our business is to produce small, light engines, how can we reduce this bank-to-bank stagger? Well, we can turn to our old books on race engines to see how this was achieved in the past. With bearing materials of relatively poor performance compared to today's engines, and large cylinders - often supercharged - we might reasonably expect that racing aero engines of the 1920s and '30s would have a large bank-to-bank stagger. The reality is that many were designed to run zero bank stagger by using one of a number of con rod designs.

One popular solution was to use a 'fork and blade' rod assembly. A conventional con rod (known as the 'blade' rod) is run in the centre of the crankpin, and a forked rod operated by the piston on the opposite bank, running two smaller bearings also run centrally on the crankpin, with its bearings running either side of those on the blade rod. There are variations on this scheme, and also other ways of achieving the same goal of zero bank stagger.

Fig. 1 - A 'fork and blade' con rod, typical of those used in old V12 piston aero engines

Written by Wayne Ward

One engine component represented such a significant development that it won the Autosport Show award for Best Technical Innovation 2011. Given that the con rod has enjoyed more than a century of development, it might be a surprise to find that this component beat many other new technologies.

Engine component design specialist Zzuhl has a technical partnership with British crank and rod specialist Arrow Precision, with Arrow producing con rods in a material in which rods have not been available before. Zzuhl has developed an ultra-high strength steel with the specific aim of producing con rods, although the material has properties that make it an attractive proposition for other engine components.

The new material, called AZZ-2 .1/M2000, has extremely high strength for a rod material, with a UTS in excess of 2100 MPa and high fatigue strength. While many materials of similarly high strength suffer from a lack of ductility, AZZ-2 .1/M2000 retains a significant degree of plastic elongation, coupled with unusually low notch sensitivity for a material of this strength.

Notch sensitivity, as the name suggests, is a measure of how a material is affected by notches and other stress-raisers such as threads and so on. High-strength materials have a higher notch sensitivity than those with lower strength, and if highly notch-sensitive materials are selected, the increase in fatigue strength in the region of a high-stress concentration is significantly derated. Therefore a combination of high strength and lower than usual notch sensitivity is a very attractive combination in a cyclically loaded component with significant stress concentrations. This is a good description of a con rod.

A number of people report that a steel con rod can be used satisfactorily without a small end bush, especially where DLC-coated piston pins are employed, and the Zzuhl rods displayed at Autosport follow this pattern. In addition to dispensing with the mass of the bush, the small end is made smaller and it no longer has to contain interference-fit loading resulting from the fit of the bush.

So, where might we see this material used in the 2011 season? Zzuhl considers this a good candidate material for high-boost applications. With the continuing trend toward highly turbocharged production engines and a raft of new race engine regulations that either allow or mandate the use of turbocharged engines, the material seems well placed to allow engine designers to produce durable rod designs.

In general, a material with an increased fatigue limit where stress concentrations exist can be used to produce lighter components by shedding mass where existing designs are strength-limited. For an existing design of con rod, this higher strength material can be used at higher load, or simply to give a greater factor of safety against failure.

Fig. 1 - Acclaimed as Best Technical Innovation at this year's Autosport Engineering Exhibition, this rod uses ultra-high strength steel (Courtesy of Zzuhl)

Written by Wayne Ward

The subject of joint shear stiffness was raised, and it was noted that the ring dowel, having a greater cross-sectional area, provides more stiffness to the joint. A stiffer joint is more stable and less likely to suffer from joint face fretting wear. While the con rod bolt may not have come undone during service - nor indeed lost an appreciable amount of pre-load - there is often evidence of small-scale movement, as shown by fretting damage.

There is therefore some merit in minimising the amount of fretting damage, or in engineering its complete prevention, and this is effectively taken care of by increasing the shear stiffness of the con rod joint. Increasing pre-load will be helpful, and providing a thicker ring dowel would also be working in the correct direction. However, this may require a heavier bolt, and more joint-face contact area; we would also need to move the bolt axis further away from the big-end axis, increasing the bending stresses on the bolt.

So, while increasing load and stiffness might be working in the correct direction to solve the problem, we will often invite further trouble by doing so. Far better that the engineer should consider practical ways to use the smallest suitable bolts and look towards geometrical solutions to the problem in hand.

For a number of years, some con rod manufacturers who make rods for production vehicles have produced 'cracked' rods. A production con rod, generally made by sintering of powdered steel, is made and a notch is deliberately introduced such that, if the rod is subject to a sudden impact in the correct location, a brittle fracture is produced running across the rod at the big end axis.

This brittle fracture surface, having failed with no elastic deformation, is complex and random, and the two halves locate together perfectly. Moreover, the method is more economical to produce than a conventional rod, as there aren't any joint-face machining operations to consider.

In producing a con rod from a high-strength steel, we are unlikely to be able to produce the same phenomenon of brittle cracking and so, if we seek to produce a similar effect, we need to do it on a larger scale by machining. There are a number of companies producing con rods who can accurately machine serrations into the joint face.

The most popular method is to produce a number of serrations in a similar form to a rack-tooth profile, running parallel to the big-end axis. These lock the rods together in one direction, but to avoid relative movement of the rods in a direction parallel to the big-end axis, a pin or dowel is still required. An example of a serrated rod split face can seen in the accompanying pictures. For the serrations to mate together great accuracy in manufacture is required.

Fig. 1 - 3D detail of serrated rod split line (Courtesy Zzuhl/Arrow Precision)

Fig. 2 - Con rod incorporating a serrated split line through the big end (Courtesy Zzuhl/Arrow Precision)

Written by Wayne Ward

It is important that these location features are machined into the rod before the big-end bore is finished to size. This guarantees that when the rod is assembled and the bolts pre-loaded to their design load, the big-end bore should take the same form as it did immediately after final machining. If the two parts of the rod are not positively located, the misalignment could cause serious problems, with bearing clearances and rod-end float likely to be affected.

There are several methods of achieving this location, one of the most common of which is the use of location pins or dowels, one on each side of the rod. Given the availability, quality and price of loose rolling elements from needle roller bearings, these parts are often used as location pins for con rods.

Another common method is to use ring dowels. As the name suggests, these are hollow dowels and are designed to fit concentrically with the big-end bolts. They are used in pairs, with one concentric with each bolt, and are generally a tight fit in either the blade or the cap, and a light push-fit in the opposite half. Practical advantages with this method are that the machining is relatively inexpensive and the tooling is quite sturdy.

These two methods are quite similar in approach, using two cylindrical components to locate the two rod halves. The ring dowel has a larger cross-sectional area compared to the smaller pins and can therefore add a little to the 'shear stiffness' of the joint face, thereby better maintaining the location of the two halves during operation.

However, one disadvantage is that the dowel necessarily moves the bolt axes further away from the big-end bore than could be achieved using dowel pins. The result is that the maximum stress in the bolt is increased in service due to increased bending loads in the bolt shank.

The minimum distance that the ring dowel causes the bolt axis to move, relative to a rod whose parts are located with pins, is equal to the wall thickness of the dowel plus the radial clearance between the bolt and the inner wall of the dowel. We can see that, in order to minimise the increased bending stress in the bolt, there is an incentive to use dowels with little radial clearance to the bolt and as thin a wall as is possible while maintaining accuracy of location.

One method that would keep the bolt axes close to the big-end axis without having to use pins or dowels would be to use a location diameter on the bolt which is a close sliding fit in both halves of the rod. However, this method is rarely used.

Fig. 1 - This con rod uses one pin on each side of the rod to maintain location between rod and cap (Courtesy of Vitesse Engineering Services)

Written by Wayne Ward

In these days of computer simulation, we can run analyses to study this phenomenon. For many years, however, engineers had to rely on formulae and a lot of painstaking calculations, as laid down in books such as "A Handbook of Torsional Vibration" by Nestorides. These calculations, which can pretty accurately predict natural frequencies and their amplitudes, are made far simpler now by the use of the ubiquitous spreadsheet.

These types of 'hand' calculations - when 'hand' literally meant writing out the formulae and working them through with a pencil and slide-rule - were a legal requirement for engine designers in certain fields, most notably where engines were designed for marine use. So a large body of literature exists on the subject, albeit probably now collecting dust on rarely visited university bookshelves.

Natural frequencies, amplitudes and stresses were calculated, and one had to prove that in use the engine either never got into the dangerous speed range concerned, or the frequency was a transient only ever encountered below cruising speed for example. These calculations had to take into account each different engine installation. Where avoiding critical speeds was not possible, boat operators had to be aware of which engine speeds or water speeds to avoid, as well as speeds that must be travelled through as quickly as possible.

Valvetrain simulation looks at the excitation of the system by the cam profile, and we aim to avoid frequencies that would see the spring resonate and surge, leading to loss of valve control.

In terms of the con rod, we also need to be aware that it has its own critical speeds, some of which may be in the operating speed of the engine. If the con rod is operated within a range close to its natural torsional frequency, the rod begins to twist back and forth.

At the point of closest approach between the valve and piston, this could cause the valve to make contact with the piston, and this is one reason why valve pockets are designed to be a reasonable amount larger than the valve. The greater volume of valve pockets leads to lower compression ratios in engines where it is difficult to achieve the desired ratio.

Depending on the level of stress, the continuous running of the engine in the region of torsional resonance of the con rod may lead to fatigue problems. Bearing problems may also occur.

As with the torsional pendulum we may have studied at school or university, where the frequency depended on the dimensions and modulus of the wire and the inertia attached to the end, the proportions of the con rod and its materials have a critical bearing on the natural frequency, as does the inertia of the piston assembly about its axis. A stiffer con rod or a lower inertia piston will tend to increase the engine speed at which the resonant frequencies occur.

Fig. 1 - It is important to avoid running constantly in resonant conditions (Courtesy of Pankl Racing Systems)

Written by Wayne Ward

In the application concerned - four-stroke, single-cylinder race engines with 'assembled' cranks, where the crankshaft isn't a single piece but is assembled with the con rod in place - there is an advantage in terms of design simplicity: the con rod can be a single piece rather than an assembly split at the big end. This means that there is neither a requirement for bolts to secure a cap, nor dowels or pins to ensure correct rod-to-cap alignment.

The con rods in question are made of metal-matrix composites (MMCs), and an early Race Engine Technology Monitor article goes into some detail regarding the properties of such materials and their application to con rods.

There are obvious advantages to using such materials for con rods, even if we can't dispense with the big-end bearing. The low mass of the con rod means transient engine response should be improved, and indeed power output should also be improved because of the lower frictional loads due to the lighter rod. There is potentially further scope for optimisation by further lightening the crankshaft based on the lighter rod.
I spoke to rod manufacturer MX Composites, from Sweden, which specialises in four-stroke motocross rods made from MMCs (see Fig. 1).

Operations and sales manager Claes Isaksson told me the company has supplied rods to many successful riders and teams; Fig. 2 shows Filip Bengtsson, who is currently challenging for the Swedish MX2 championship using MMC rods.

MMC properties are due to the addition of small particles of silicon carbide to the aluminium matrix. These additions, however, have the unpopular side-effect that the material becomes difficult to machine. This is partly why some of the more popular rods are made from near net-shape forgings.

The machining of the material is also a very specialised process, using special tooling and a high-speed machining technique originally developed by Saab. Isaksson says, "High-speed machining is one of the keystones in this process, as are the special tools being used."

In many ways motocross is an ideal arena in which to test a rod without a bearing. The engines are simple but highly tuned, and are stripped and rebuilt regularly, giving ample opportunity to gauge the condition of the internals. The move away from the standard needle roller came because it was felt that the original big end of the rod was rather large, so there was a desire to see if the needle roller was really required. From the work undertaken, it appears that there are cases where the needle-roller may be dispensed with.

With these motocross rods running a hydrodynamic bearing, but without a shell, how long will it be before we see this technology more widely applied?

Fig. 1 - MMC rods are mechanically simple but require special machinery and tooling for production (Courtesy of MX Composites)

Fig. 2 - Swedish MX2 contender Filip Bengtsson uses MMC rods to good effect (Courtesy of MX Composites)

Written by Wayne Ward

Despite the fact that these high-precision parts are widely available at quite reasonable cost, there have been a number of attempts to dispense with the bearing shell in recent times, and some of these have been reported in both Race Engine Technology magazine and RET-Monitor. One materials article in Monitor discusses the fact that one company offers a product line in metal matrix composite (MMC) rods for four-stroke applications which run successfully without any form of bearing. In this application the original bearing was a needle roller, but the fact that the engine has run successfully with no bearing shows the potential to do so on a wider basis.

For the motocross application detailed in the article above, there has been no attempt to substitute the bearing with anything else, and it is in this respect that it differs from other attempts. I have spoken with a number of industry experts on the possibilities for running without a bearing, and many have made the point that the success of this approach would rely on having excellent oil filtration. An important aspect of conventional shell-bearing technology is the ability of the soft metal platings to trap small, hard particles so that they don't cause damage to the crankshaft or the big-end bore. In dispensing with any form of bearing shell, we run the risk of hard particles passing through the journal bearing and causing abrasive wear damage.

An obvious solution, although technically difficult, is to apply the soft metal platings directly to the bore of the con rod. This approach is being developed by one European coating supplier, and its solution seems advanced. It claims to be able to coat many different materials, with some others not being tried as yet. But it can successfully coat both steel and titanium.

There are various advantages to being able to dispense with the bearing shells. Clearly there is an advantage in terms of reduced mass, but there are other significant advantages. By being able to place the rod bolts closer to the applied load means the bending stresses in the bolts will be lower. Moving the bolts closer to the crankpin axis also has positive implications in terms of a smaller rod locus. And the tensile stresses in the rod due to the interference of the bearing shell are no longer present.

While this technology is far from maturity, it will surely begin to make an impact on the design of both series production and racing rods in the near future. For the time being, only a very few rods will be able to run successfully without some sort of big-end bearing.

Fig. 1 - In future will racing rods such as this feature a coated big-end bore, rather than using a bearing shell? (Courtesy of Carrillo)

Written by Wayne Ward

The oil entering the small end via these radial drillings has to find its way into the entry of the drilling (hence the generous chamfer provided), and the source of the oil may in some cases be only oil 'mist' as the result of splash lubrication. In racing engines with dry sumps and higher power outputs, it is quite likely that piston spray jets will be provided to take heat from the piston. In this case there will be a lot of oil in the vicinity of the small end.

Sometimes this will be a single spray jet, but some modern racing engines can have up to ten spray jets per piston, some of which are aimed specifically toward the small end of the rod. The recent trend towards stiffer pistons and the 'box-bridge' design, however, means there is some obstruction to oil flow aimed in the vicinity of the small end.

In this case, and others where there are extreme bearing stresses, there is a case for providing positive lubrication of the small end - that is, a specific and pressurised oil feed. Highly supercharged engines are a good example of an application where positive lubrication is often provided.

The oil flow to the small end is via a drilling from the big end of the rod. The photo here shows a connecting rod that has a drilling for this purpose.

Oil arriving at the crankpin via the crankshaft oil drillings is fed onto the outside of the bearing shell via a drilling through the shell itself. The hole may connect directly with the drilling up through the connecting rod, in which case the hole in the shell will be vertically upward. Some people are wary of having a drilling at this position in the big end shell, however, so they might therefore choose to provide a slot or cavity to carry oil to the drilling from a hole which they perceive or calculate to be at a more advantageous point in the shell.

If you choose this route you have to make sure the bearing itself has enough support, especially if it is a thin bearing. The trend toward thinner big-end bearing shells makes it more important to support the bearing in a stiff housing.

If you're designing a bespoke rod you may have to carry out a lot of FEA analysis or physical design iterations before arriving at a suitably stiff housing design. Taking away material from the surface of the housing means this part of the bearing may deflect under load and therefore not play its proper part in supporting the forces in question. This leads to higher bearing stresses on the rest of the bearing and therefore may overload these areas, causing premature failure.

Fig. 1 - A con-rod with a drilling from the big end to provide oil flow

Written by Wayne Ward

As was mentioned in the early RET Monitor article on the subject of the little end and its design features, it is thought possible by some experts even to dispense with the bronze bush provided in the small end in some circumstances. It should be noted that this is not possible without taking other measures to ensure that the bare small end bore is suitably finished and of the correct mechanical properties, and that the piston pin is correctly designed and specified.

However, we must provide some lubrication in this area, and it is certainly possible to supply too little lubrication. It has been traditionally the case in the past to design con rods to have a single hole through the top of the small end bore into which oil, being carried around in the turbulent motion of the air / oil / blow-by mist in the crankcase could find its way. As engines have become more powerful and loads and pressures have increased, we haven't really seen an equivalent increase in the dimension of the piston pin, either in diameter or supported length in the con rod or piston. The same is true also of the series production vehicle engines, where 'power density' (the ratio of power to engine mass) is increasing as we seek to provide sufficient performance to our cars with a lower engine and vehicle mass. As we have increased the load applied to the small end contact and the frequency of oscillation we have seen a trend to increasing the number of holes from one to two (or perhaps more in some cases) and this has necessarily meant that these are no longer on the centre-line of the con rod. At this point we should take note of the con rod designs from two-stroke engines. The accompanying picture shows the small end of con rod from a production Honda motocross engine. The pattern of the two holes shown here is very similar to that now used on four-stroke competition con rods in many racing categories.

Referring to our mechanical engineering text books and taking to pencil and paper calculations can point us in the correct direction for the best (or indeed worst) positions for these holes in terms of stress. Clearly, finite-element stress analysis would be a further step toward accurate calculation of the effect of these holes on the stress field around the small end, if used correctly.

Fig. 1 - This Honda CR250 con rod has two holes allowing oil to enter the small end

Written by Wayne Ward

Previous articles on the subject of con rods have talked about some of the material choices for these parts, and in the recent magazine Focus article on the subject of con rods, the author discussed the merits of material choice with leading manufacturers of con rods from all over the world. There is a general feeling that reduction of mass and inertia is a good thing, and in very many cases this is true and the title of this article can be taken almost as a 'universal truth' within the limits of maintaining both satisfactory function and sufficient reliability.

For those chasing light weight, low inertia and reduced mechanical loads, there are indeed significant gains to be had from working on the con rod. Light con rods make less demand for counterweighting on the crankshaft and so the any mass saved on the con rod can be carried on through the design of the whole cranktrain. In many forms of motor-racing this is the philosophy that is followed with good results. Low inertia is felt to improve the responsiveness of the engine, making it 'pick-up' quickly when transferring from a closed or neutral throttle to an open one. In a similar vein, low inertia turbine and compressor wheels in turbocharged applications prevent excessive lag and improve responsiveness.

So, are there any racing applications where low inertia and low mass are not primary goals? The answer to this question in a surprising number of motorsport situations is 'yes'. Whilst we can usually be certain that riders / drivers / pilots of racing motorcycles, cars and boats will want more power than their rivals, there is an aspect to performance which is gaining increasing understanding, but is little understood by many (probably most) involved in racing engine design.

When we talk of performance, those involved in engine design alone will generally judge peak power to be the metric by which performance is measured. Those who are perhaps more enlightened may have other methods by which they judge performance. Ideally we ought to judge performance by the stopwatch, because I suspect that there are very few of us who compete on dynamometers alone for any sort of peak power prize (although I have seen an extremely cheap in-house 'Power Cup' awarded to the builder of the most powerful engine) and so our efforts are judged on the track.

To be of greatest use on track, even the most power-hungry engine supplier realises that his product needs to have a certain useful operating range, and this is a step in the correct direction and this represents a loose grasp of the subject of 'driveability' or 'rideability', which goes beyond providing a high output.

There are a number of applications where low inertia is actually a barrier to performance as measured by the stop-watch. In some of these applications we have, by constantly producing parts of lower inertia, not provided what is required. In some instances, we have actually seen titanium con rods being replaced with steel rods, and furthermore the steel rods are specially designed for increased inertia, in order to control engine response. These have proven repeatedly to offer improved performance when measured in terms of lap time and race times. Such con rods appear ungainly to the eye when judged against 'normal' racing con rod designs. However, they do represent a measurable improvement in performance, and so should be judged on their considerable technical and economic merit. Providing a similar gain in laptime by increasing power alone would cost a great deal more in spent treasure.

Fig. 1 - Steel rods may offer real performance gains compared to lighter titanium rods in some applications.

Written by Wayne Ward

Photo courtesy of Auto-Verdi.

Con rods are a particularly suitable application, and most companies who manufacture racing rods for 'serious' applications, i.e. where the rods are significantly stressed, will shot-peen their rods as a matter of course. The only application that I have found where there is an exception to this rule is in drag-racing where aluminium rods are often used. For titanium and steel con rods though, it is common to shot-peen and rather than use the services of specialist subcontractors, some rod manufacturers have their own facilities. Components which are subject to simple tensile loads are not greatly affected one way or the other by residual compressive stresses, but those subject to bending and torsional loads are aided tremendously in terms increased fatigue life by processes which impart these stresses. Con rods are indeed subject to significant bending loads and also some degree of torsional loading.

A few years ago I was shown the remains of an engine, modified for motorcycle racing by an amateur racer. The particular wizard in question had apparently taken exception to the 'flashing' left by the forging process on the production rods, and had taken his trusty grinder to the rods to remove this material. He had left the rod marginally smoother than a normal production rod, but had left some significant machining marks perpendicular to the load path, which is the about the worst possible damage he could have inflicted with his grinder. Unsurprisingly, one of the connecting rods had failed. I have no doubt that this amateur 'tuner' had no knowledge of shot-peening, otherwise he would have treated the modified rods by this process.

It is, to some people, counter-intuitive to shot-peen rods, or any other component, even to students of engineering and newly-qualified engineers. In studying cyclic loading and the phenomenon of fatigue, students learn that improvements in surface finish lead to an improvement of fatigue strength. They learn that a machined finish is better than a cast, or as-forged surface finish, that a ground surface finish has better fatigue properties than a machined surface, and that a mirror polished surface has even greater fatigue strength. In this regard, we might not be able to fault the logic of the previously-mentioned 'tuning wizard'. However, it is very likely that the rod had previously been shot-peened, and that its surface was in a state of significant compressive stress. By introducing grinding marks perpendicular to the main load path, he had made one serious mistake, and by removing the compressively stressed material and not thinking to restore it by shot-peening, he had made another. Unfortunately for my friend, who had purchased the result of this grinding fanatic, the outcome for his wallet was not a good one.

Fig. 1 - Shot-peening can prevent expensive connecting rod failures.

Written by Wayne Ward

The reliability, function and precision of these bearing shells rely greatly on the quality and accuracy of the bearing housing, its correct design and manufacture and method of bearing retention.

In a two stroke con rod, which invariably runs rolling-element bearings, con rod manufacture is relatively simple as the bearing bore is finished without having to consider the split which is a feature of the four-stroke con rod.

However, when we consider a four-stroke engine running bearing shells, the con rod is machined with a ‘race-track’ shaped (more like Homestead than Spa!) bearing bore before the rod is split. Once the rod has been split by machining or other means, the two halves are clamped together again, the ‘straights’ on the racetrack shape having been removed as part of the splitting process. The load in the bolted joint at this point is critical; it must be the same as will be used when the con rod is installed. If it is significantly different, the bearing bore will be distorted in service. Once the rod is bolted, final machining of the bore is completed, by precisely boring to size, and perhaps a honing operation.

There are other design features to consider within the big end bearing bore, concerning clearance to adjacent components, bearing retention and lubrication, but more of these at a later date.

One point that we need to consider is that of stiffness of the bearing housing in the con rod. If there is insufficient material around the bearing bore to properly support the bearing in operation, maintaining its position and form, the bearing will simply deflect and parts of its surface will not be able to support its fair share of the applied load. In this case where the bearing support structure is of insufficient stiffness, premature bearing wear and hence failure is likely in highly-loaded situations. Bearing shells in con rods with oversized lubrication grooves machined into them, or where weight-saving has been practised with too much zeal, show clear signs of increased wear in areas where bearing support is insufficient.

The big end bearing area is an obvious target for weight saving. Examination of any con rod will quickly show that, given the amount of material in this area of the con rod, any saving made here will probably be greater than can be achieved elsewhere on the rod. The saving of mass at the big end is doubly potent in that any saving made here is mass that does not have to be considered when designing the crankshaft counterweighting. So, as with so many design decisions, we must balance our requirements for minimum mass with the need to provide sufficient support to the all-important bearing shells. This is not a simple matter when pushing hard for minimum engine mass, and amateur tuners and Formula One engine manufacturers alike make the wrong judgement.

The independent rod suppliers have a good view on what works and what is a step too far in terms of mass saving, as they tend to see a great number of rod designs and have, through experience, come to know what is likely to cause problems. One such supplier, Arrow Precision, who I visited recently, have designed hundreds of different con rods, and it is such companies whose advice and experience is invaluable to those without access to powerful FEA simulation tools.

Written by Wayne Ward.

So, is there a case for steel as a con rod material, other than on economic grounds or lack of familiarity with new materials? The answer is, in some cases, yes. We have previously looked at some of the drawbacks of titanium; it can be said that part of the continued attraction of steel is due to the disadvantages of titanium.

Certainly steel does not suffer from any of the surface related woes of titanium. Galling, the process whereby cold-welding of material causes seizure, does not affect steel, but can be a serious problem for titanium. Galling can affect threads and so the installation of con rod bolts can be very sensitive to this effect in titanium con rods. We know that the con rod bolt is one of the most highly-stressed components in a racing engine, and therefore one of the most critical. Failure of a con rod bolt often spells disaster in a racing engine, resulting in extensive (and expensive) damage. The other area in which titanium can suffer is the small end. Titanium con rods can suffer galling during insertion of the small end bush and this galling damage can act as a stress raiser. Steel does not suffer this problem and so there is no need to worry when inserting a bush, or when re-bushing con rods. Owing to the fact that the surface of titanium is liable to be damaged even under low levels of stress, titanium con rods have to be carefully looked after and handled. Steel con rods are not so sensitive in this regard.

There are some advantages to using steel, and these are not to be underestimated. The stiffness of steel is much higher than that of titanium and this can be of use when designing the con rod. It is also of direct use in mitigating the level of cyclic stress in the con rod fastener. As we have illustrated in the print issue of Race Engine Technology (RET 41, September/October 2009) the ratio of bolt stiffness to joint stiffness, along with the service loads, dictates the level of stress amplitude in the fastener. In this context, a higher stiffness for the joint components is advantageous.

The disadvantage of steel, in comparison to all of the other ‘main contenders’, is its density. With a density of 7.85 g/cc, it is almost 80% more dense than a typical titanium con rod alloy. This is a serious deficit in most circumstances. However, in an attempt to calm power delivery, some race engine developers have started to look seriously at steel because it offers a way to increase the inertia of the engine. People have been putting steel con rods into engines originally equipped with titanium rods purely for reasons of controlling power delivery.

Written by Wayne Ward.

One point that was covered last month was the effect that the use of titanium has on the proportion of the externally applied loads that the con rod bolts experience. This is affected not only by the magnitude of the external loads (in this case, we are really talking of the ‘inertia’ loads that we see at TDC on the exhaust stroke on a four-stroke engine), but by the material choices and geometry of the fastener and con rod. The proportion of the external load experienced by the fastener is controlled by the ‘load coefficient’ which is influenced by the stiffness of the con rod and the bolt. If we use a less dense con rod, and design it properly, we can expect that the external loads will be higher. However, in choosing titanium as a con rod material, we increase the load coefficient, thereby increasing the proportion of the external load that the con rod bolt will have to cope with. We must be careful of this point when choosing our con rod material.

However, the low modulus of titanium in comparison with the commonly-used bolt materials means that the distribution of the load along the threads is much more even than with a steel con rod. For metric threads and identical nut and bolt materials, around 85% of the load is taken by the first four threads, in a nut with ten threads. If the nut material (in this case the ‘nut’ is the con rod) is of a lower modulus than the bolt, this situation is much improved with a lower proportion of the load being taken by the first thread, thereby causing a much lower stress concentration at this point.

One material which we have heard much of in connection with valves, and comparatively little regarding con rods are the titanium aluminide (TiAl) materials. Now banned in Formula One under rule 5.13.1(c) which bans all intermetallic materials (or which titanium aluminide is one), which are defined as having an intermetallic content above 50% by volume. If you have read it, clearly the rule was written by a materials expert, or with the aid of one, but it hasn’t stopped titanium aluminide materials being developed which have close to, but less than 50% of intermetallics. These materials are aimed at use as valves and con rods mainly in Formula One (when a new set of engine rules finally arrives). However outside of Formula One, where technical rules are less restrictive, titanium aluminide materials have been successfully used as con rods in racing conditions as was confirmed when I recently spoke to an engineer from the company involved in the design and manufacture of these parts.

In comparison with titanium, TiAl has a lower density (less than 4 g/cc) and higher modulus (varies from around 160 GPa to around 175 GPa). In fact, commercially available TiAl materials, which are being considered for possible series production parts for the automotive industry in the long term, not only contravene the above Formula One regulation, but also the specific modulus regulation which imposes a maximum ratio of elastic modulus to density of 40 GPa/(g/cc). It is a shame that Formula One is not in a position to develop materials technology which will benefit road car engines by making them more efficient. The advantages of lightweight rotating and reciprocating components in terms of reduced friction are well-known.

Written by Wayne Ward.

Titanium has long been used as a con rod material for racing, and its benefits have been slowly taken up by road car and motorcycle manufacturers in the last couple of decades, although this has only been on very high-specification machines, many of which will have found their way onto race circuits. That great racing company, Honda, pioneered the use of titanium con rods for road vehicles in the late 1980s, first with the fantastic RC30 motorcycle, and then with the NSX road car.

Titanium is favoured by many owing to its low density. A typical titanium con rod material will have a density of approximately 4.4 g/cc, compared to 7.8 g/cc for steel. If we were simply to copy a steel con rod and produce it in titanium (many people do just that), we would realise a 44% mass reduction. This alone is an extremely worthwhile saving, but the benefits go much further than this.

Any mass saving on the con rod should, if we have the resources to properly take advantage of it, be accompanied by a consequent mass saving on the crankshaft, where we add counterweighting to balance the rotating and reciprocating masses. These counterweights can thus be made lighter and so we find that any mass saving made on the con rod is effectively doubled if we reduce the crankshaft weighting accordingly. An extra benefit is reduced inertia which allows the engine to accelerate faster, giving a performance benefit to the car/motorcycle, in spite of the fact that the engine may not produce any more power. However, we should note that this may make the engine feel more ‘peaky’ or ‘highly-strung’, and this is not always what is required in some forms of motorsport, especially if traction is limited, or ways of controlling it are banned.

In common with aluminium as a con rod material, some engineers state that the elastic modulus of titanium compared to steel makes it a more forgiving material and it can therefore have a valuable role to play in protecting adjacent parts of the engine from shock loads due to, for example, irregular combustion (detonation). The elastic modulus of Ti-6Al-4V, a popular titanium alloy for con rods, is 114GPa, compared to 207GPa for a typical steel (the specific modulii of titanium and steel are very similar).

We should say that, in order to optimise the material for use in con rods, we should not simply copy a steel rod, because in doing so we miss out on some of the benefit of using titanium. If we design the small end to have acceptable stresses, we will find that it is lighter than the equivalent steel item, and therefore each part of the rod between here and the small end is consequently less stressed under maximum inertia loading than its steel counterpart. We may also find that we are able to use a smaller con rod bolt, although the ‘load coefficient’ which dictates the share of cyclic loads between the bolt and joint is changed compared to a joint of the same geometry in a steel rod.

So, what are the disadvantages of titanium as a rod material? Cost is obvious; it is still much more expensive than steel. Quality might be a concern, especially if we are tempted to buy the very cheapest material available, and I’ve seen some appalling material quality. Tribological behaviour is poor, with galling a serious concern with fasteners and interference-fit bushes, although these problems can be overcome with experience.

Written by Wayne Ward.

In terms of circuit racing, the main types of materials used are steel and titanium but in drag racing, we find that aluminium is widely used, especially for the high-power classes such as top-fuel. We shall begin our ‘tour’ of rod materials by looking at aluminium.

The extremely successful US Army car of Tony Schumacher is equipped with forged aluminium con rods. Aluminium was widely used for automotive con rods in the past, but its use has been almost universally supplanted by steel for production engines. So, what makes aluminium a good choice for a drag-racing con rod, especially given the huge forces involved? Recently, we have canvassed the opinions of a number of con rod manufacturers for a forthcoming article in the magazine, and these included those making aluminium rods for top-fuel.

When we look at the properties of aluminium, we find that, in comparison to steel, that it lacks both strength and stiffness. Furthermore, when we look in greater depth, specifically at the fatigue properties of the materials, we find that aluminium has no defined fatigue limit. In terms of looking at a S-N curve (Wohler diagram), there is no knee in the curve at the high cycle end. Whilst the gradient of the S-N curve is shallow, it is not flat, and this means that, whatever the level of stress-amplitude, an aluminium part will always fail eventually, whereas a steel part will not, providing that the stresses are within certain limits.

The strength of the rod material is not such a concern, as the specific strength (strength divided by density) is not so different from steel materials, so that in respecting the particular level of material strength, the rod should be around the same mass. For those that are involved in the design and analysis of con rods, the nominal stress levels in the shank of the rod are often not very high. So, on the face of it, providing that the rod is properly managed in terms of service life (mileage or time limitations), we could possibly make a lighter rod in aluminium.

However, the main reason given for the selection of aluminium for drag-racing rods is the ability of the material to absorb shock-loads. In this case the rod is acting elastically as a spring, albeit quite a stiff one. It is here that we find that the properties of aluminium are used. Its low modulus of elasticity means that the rod can be designed to have a low stiffness and thus act as a cushion for protecting other parts of the engine (piston and bottom end). Drag-racing can have a lot of detonation in the combustion and the resulting high load spikes can be very damaging to the engine components. So, the rod needs to be designed to an appropriate level of stiffness to suit this. Whilst this can be done in any material, if we were to do this in steel, we would find that the rod has a much smaller cross-sectional area and, more importantly, a lower second moment of area (also known as moment of inertia, area moment of inertia or second moment of inertia), and would therefore be more susceptible to suffer failure due to bucking under abnormal loads.

Written by Wayne Ward.

There may be some among you who aren’t familiar with the workings of a cyclically loaded joint and who don’t possess a good textbook which covers the matter, but we can say that the greater the ratio of stiffness of the clamped members to the clamping member (i.e. the bolt), the smaller the proportion of the cyclic load that the bolt will see. This will be covered in greater detail in articles on fasteners, but it is worth knowing this useful morsel when considering the design of the joint. It would be quite easy to design a con rod assembly with the rod and the cap closely sculpted around the bolt head and shank for minimum mass, but this is a sure way to disaster where significant cyclic loads are present. As can be seen from the picture accompanying the previous article on big-end design, the joint is deliberately made quite stout in the area of the joint face.

By keeping a high stiffness ratio as described, we can use a smaller bolt, which in turn has benefits for the rod assembly, the rest of the engine and ultimately the performance of the car. The smaller bolt is, of course, lighter and, being smaller, can be placed closer to the big end axis which has two main benefits. The first benefit, which is obvious to the naked eye, is that the rod becomes narrower. On vee-engines, the width of the rod at this point can become the limiting factor in lowering the crankshaft axis in the engine. A lower crankshaft height allows packaging and performance benefits in the car, not least from having a lower centre of gravity for the engine. Prior to the Formula One engine homologation rules, it was not unknown to have the sump machined locally to provide minimum clearance to the rod locus; the lowest point of which was defined by the big-end bolt. Another recently used solution to this problem was to use a big-end joint which is angled. The cap is still applied axially to the rod, but each joint face is angled, with the outer edge above the nominal centreline, thus angling the fasteners and providing more latitude to lower the crankshaft. Of course, the edge of the joint at the inner edge must still be on the nominal joint centreline.

The second good reason for keeping the bolt axis close to the big end axis is the reduction of bending load in the bolt. As is obvious from examining the con rod, the load is not applied to the joint in a simple axial fashion, but is a combination of axial and bending loads. To mitigate the effects of this bending stress, it is possible to make design changes to joint face, involving judicious removal of material from the mating faces. Another factor which can allow the fasteners to be placed closer to the rod axis is by not using a ring-dowel around the bolt, although this remains a popular solution to the problem of locating the cap accurately with respect to the rod. However, this can be achieved with a pin dowel and good-quality small diameter individual rollers as found in needle roller bearings, often used for this purpose. An alternative is to use a bolt with a tight-tolerance shoulder only slightly larger than the bolt thread.

Written by Wayne Ward.

One of the main jobs that the big end has to do is to house and retain the bearing which is almost exclusively of the plain journal type made of two individual halves. The most popular method by which these are retained is by the use of ‘tags’ which are pressed or stamped into the bearings. Each part of the con rod (which we shall refer to as the cap and the blade) has a corresponding machined slot. An alternative method is to pin the bearing but, given the thickness of the big end bearing shells used currently, this method of bearing retention lends itself much better to the main bearings in the cylinder block. I have also seen con rod designs which rely solely on the heavy interference of the bearing in the big end. However, I don’t know if these have been successfully raced.

In Formula One, where the bearings are very thin, attention has to be paid to the stiffness of the material around the bearing such that it provides sufficient support to the bearing. If the bearing is not adequately supported by the surrounding material it tends to push away under load and thus overloads the better supported parts of the bearing leading to premature wear and failure.

If the design philosophy of the con rod requires that the small end is positively lubricated, oil is supplied via a drilling between the two ends of the con rod. Clearly the oil feed to the drilling must be via a hole in the bearing and the location of this hole is, in general, not directly in line with the drilling between the big and small ends of the con rod. In this case, the drilling is supplied via a slot machined into the bearing surface of the con rod blade which extends between the drilling and the feed holes in the bearing. It is clear that the bearing is unsupported where such slots are provided and this has to be taken into account when sizing the bearing and designing the con rod.

In designing the con rod, there are various different ideas applied to the design of the con rod cap, and these employ polar opposite philosophies. It is popular to provide quite a deep cap which is very stiff with the aim of maintaining a circular bearing housing. The stiffness of the cap comes at the expense of increased mass which consequently leads to increased rotating crank assembly inertia. The mass and stiffness are the subject of a trade off to arrive at an acceptable compromise. An alternative is to provide a deliberately flexible cap, where one is resigned to the fact that the cap will deflect and wrap around the crankpin thereby limiting the deflection of the conrod. This is less popular than the stiff cap philosophy, but the benefit of this flexible design approach is reduced mass and inertia.

Next month we shall look at the remaining aspect of the big end – the joint face design and the joint itself.

Written by Wayne Ward.

The small end must provide sufficient bearing area to transmit the forces due to combustion and also to withstand the forces due to rapid deceleration of the piston assembly at top dead centre. The bearing is commonly a bush of a bronze material which is a shrink-fit in the small end of the con rod. The bearing needs to be lubricated and the bush therefore often carries oil grooves to facilitate the flow of oil around the piston pin. There are various common designs of grooves from single and double circumferential grooves (sometimes also combined with axial grooves) to helical grooves. These grooves are often machined in after the bush has been fitted to the con rod. Some Formula One con rods use coated steel bushes in place of a bronze bush.

The fitting of the bush itself may cause problems, especially in titanium con rods and a number of surface treatment processes have been used on both the bush and the con rod in an attempt to overcome the problems of galling as the bush is fitted. The stress caused by the interference fit of the bush must be taken into account when calculating the stresses around the small end. By careful design, Formula One engineers have largely been able to overcome the inherent problems in this area, but other con rod materials have been considered recently in Formula One, although they may not be currently used as they were not developed sufficiently before the homologation rules came into force.

The con rod has another role to play where the con rod is ‘piston-guided’. The con rod may be constrained from moving axially (parallel to the crankshaft axis) by either thrust shoulders on the crankpin (crank-guided con rods), or by thrusting against the inside of the piston-pin bosses (piston-guided con rods). A common design feature where piston guided con rods are used is for the usual single-piece bush to be replaced with two flanged bushes, where the bushes provide the thrust bearing surface against the piston. These bush flanges are provided with machined features to encourage the flow of oil into the contact area and to generate pressure in the oil film. The gap between the two bushes fulfils the role of the circumferential groove discussed above.

After all the talk of bushes, we should mention the work that has been done in trying to dispense with these components. At least one coating specialist thinks that they are close to being able to provide a satisfactory bearing surface in a con rod without the use of a bush. As mentioned before, the small end needs to be made larger owing to stresses due to interference fitting of the bush and the effect of galling which effectively introduces a stress concentration. An alternative way to dispense with the small end bush is to use an alternative material for the con rod, other than titanium. One Formula One manufacturer was very close to introducing a con rod designed in such a material before the engine homologation which would have run with a coated pin, with the parts having been successfully tested on the dyno.

The need for positive lubrication (a pressurised oil-feed via the crankpin oil feed) of the small end of the con rod has been a point which has been debated for some time and in both Formula One and other formulae, teams have run successfully with con rods, with and without positive lubrication. A common design feature is to have one or more drillings radially inwards from the area around the top of the con rod to communicate with the previously mentioned oil grooves. Of course, these features do not necessarily have to be plain holes….

The stiffness of the small end and the condition of the end of the hole in the bush can have an effect on the life of both the piston pin and the bush, and profiled bores are quite common in the bore of the small end bush to ensure that contact stresses at these heavily loaded areas are controlled to avoid fatigue damage.