QUESTION: In your Ignition section in the March issue, you mention that it takes 12 to 15 weeks to produce a finished crankshaft for the Yamaha M1 MotoGP engine. It would be fascinating to have Kevin Cameron explain why it takes so long.
Scott Saunders
Binghamton, New York
ANSWER: At first it seems strange that it would take so long to build the one-piece plain-bearing crankshaft of a MotoGP machine because we have all seen YouTube videos of CNC machines milling and turning a crankshaft out of a solid billet of steel. Done in one shift, right? Then, bang, into the heat-treatment oven, finish-grind the journals, and into the build rack.
The shape could be done in a day, were it not for residual stress. Machine tools do not actually cut. They shear, by applying concentrated force to a very small area. The metal flows, stretches, and breaks as the tool edge advances. That flowing and stretching leave highly stressed zones in the part. If the part were put into service with those stresses in place, the stress of operation would be added to that residual stress. The most-stretched interatomic bonds would then break under operating stress, and in time cracks would form and propagate.
Therefore, the part must be stress-relieved before it goes into service. The basic idea is to bring the part up to a temperature that will set its atoms into fairly vigorous thermal vibration, with the result that instead of breaking stretched interatomic bonds, those bonds are allowed to “jiggle their way” to a lower-energy state. The stretched bonds, agitated by thermal vibration, relax.
Now another problem: The hotter we make the part, the more we may drive processes of metallurgical change that we don't want—for example, the migration of carbides into the intergranular zones between the jumble of metal crystals that make up the part. But the lower the temperature of the stress-relief process, the longer it takes. Days and days.
We have some idea of elementary heat-treating. Heat up a simple carbon steel part to put its alloying elements into solution, then suddenly reduce its temperature (by quenching in water or oil), thereby allowing only tiny crystals and tiny iron carbide particles to form. In this state, the steel has maximal hardness, but is brittle. To toughen it, the part is heated to some intermediate temperature, allowing crystals to grow but sacrificing some of the hardness-inducing carbides.
Modern alloys are more complicated than this, and so require more heat-treating steps to put them into the desired condition. That takes more time.
And there's another thing: All steels have some mineral content as a result of the refractory lining of process equipment. To achieve the very highest fatigue strength (as in racing valve springs, connecting rods, and crankshafts) the steel must be processed in vacuum to allow such contaminants to evaporate and be carried away by the vacuum system. This adds further complication and takes more time. Even in VIM/VAR vacuum-processed steels of the highest quality, some tiny “stringers” (elongated mineral contaminants) remain. Heating the steel to a particular temperature and then deforming it stretches the stringers until they break, and their material coalesces into tiny droplets that are too small to act as crack nucleation sites.
The weakest part of every crankshaft is the fillet radius that joins the journals (main or rod) to the cheeks. Cracks need tension to propagate, so tension is eliminated from these regions by rolling, shot-peening, or otherwise placing the metal surface of the fillet in compression. Many times in the past, even when generous fillets are provided and placed in compression this way, crank operation still produces fillet cracking. Then the usual (and partial) remedy is to narrow the journal to allow fillets of larger radius, or to undercut larger fillets into the crank cheeks. One of the processes used to produce surface compression is to accelerate nitrogen ions into it, where they form hard iron nitrides. Jamming them into the surface puts it into compression.
Aware motorcyclists know that the trend in lubrication is toward ever-lower viscosity (thinner) oils. Yet even in production engines, the minimum oil film thickness in a loaded plain bearing can be less than 2 microns. Any surface roughness near that value will cause bearing-to-journal contact. The more you try to save power by using thinner oil, the more perfect the surface finish of the journal must be to prevent metal-to-metal contact. Not just shiny—because a surface can be shiny and still have waves in it. It has to be truly cylindrical, with the lowest possible surface roughness.
A process for achieving this was developed by Chrysler during WW II, when anything that would shorten the break-in of aircraft engines saved critically needed aviation gas. The Chrysler process employed shaped laps, floating on a film of oil, enclosing a crank journal. The hotter the oil, the thinner the film became, until the lap began to touch only the tallest surface features on the journal. By careful control, finishes of almost any desired smoothness could be achieved. But this takes time to achieve, in a shop with the necessary specialized equipment. When stock-car racer Junior Johnson decided years ago to seek power savings from lower oil viscosity, he learned that no one in the US was offering the process any more. He had to go to Germany to get his crank journals finished. Since then, the process has been widely adopted in automotive practice.
And so, the bottom line: Dorna, reasoning that the cheaper racing can be made, the more teams will rush to join the series, has decreed that each factory rider can use no more than five engines per season. There! That should cut costs a bunch! But that leaves the factories to find ways to make every part last 1,000-1,500 miles, instead of the 150-300 miles before the 5-engine rule. Achieving the necessary durability is not cheap, not easy, not quick.
Send your “Ask Kevin” questions to cwservice@cycleworld.com. We cannot guarantee a reply to every inquiry.