Creep

CREEP

IGNITION

TDC

THE UNPOETIC MOVEMENT OF HOT METAL

KEVIN CAMERON

In the later 1960s, I could see that some Triumph air-cooled 500 twins started AMA National races strongly and finished them sounding a bit flat and emitting smoke. Then in 1968, they were swept into obsolescence by, of all things, Harley’s resurgent (and supposedly obsolete) side-valve KR.

In the fullness of time, I suspect that what I was seeing was very likely the result of the slow heat distortion of solids far below their melting point, known as “creep.”

The classic example of creep usually given is the gradual elongation of jet engine turbine blades, constantly loaded in tension by their rapid rotation and kept at high temperature by the passage of hot air-diluted combustion gas among them. At installation, the blades clear the inside of their casing by enough to allow for heat expansion and some elongation. As blades grew longer in service from total-time-at-temperature, they would eventually begin to scrape against the casing if not replaced. Another such example is the movement of glaciers, which although solid are able to deform and flow under the urging of gravity.

In the early days of motoring, it was common for the heads of exhaust valves to gradually deform into a tulip shape. This, too, occurred well below the melting point of the valve material. A popular exhaust valve material in that time was tungsten tool steel, noted for its ability to retain cutting hardness even at redhot heat. Yet despite this extraordinary hot-hardness, it could creep under the steady pull of the valve spring.

Creep occurs basically because atoms in the material are able to gradually diffuse away from positions of high-bond stress to lower-energy positions. This process allows the crystals of which all metals (and ice) are made to slowly become longer in the direction of applied stress. Thus creep produces a slow and directional change of dimension. Remember that energy is not uniformly distributed throughout a solid, even though all of its atoms are vibrating within the elasticity of the electrical bonds linking them to neighboring atoms. The intensity level of this vibration is temperature.

The number of atoms per unit time that may accidentally acquire enough energy to break those bonds, move to new positions, and form new bonds is a matter of statistics and temperature; most atoms in a solid have close to the average energy, but a few with either more or less energy also exist. This is a slow atom-by-atom process, but, understandably, the hotter the material becomes, the greater the number of such “cut-and-run” atoms there will be and the faster creep takes place.

Those of us who had a relative in the Army Air Force in WWII have probably seen photos of gangs of shirtless men, pulling the propeller blades of a large piston engine around before start-up.

The usual reason for this pull-through was to be sure no cylinder had accumulated enough drained-back lube oil to hydro-lock when turned by the starter, possibly bending a link rod. If a lock was found, pulling a spark plug and draining the oil was the effective fix.

But in the case of especially hotrunning engines, there was another reason for the pull-through. That was to try to detect an “unseated” exhaust valve—one whose valve seat had become distorted by creep enough to make it leak. Once a valve begins to leak, instead of cooling rapidly in full contact with the cooler valve seat, it is made hotter by the leakage of hot combustion gas. This quickly erodes the valve, which begins to look as though it had been burned away by an acetylene torch.

Okay, that was what, a lifetime ago? How is that relevant to engines today? In talking with a veteran racing engineer who managed one of the BMW Boxer Cup bikes in races run on AMA National weekends a decade or so ago, I learned that it was essential to re-seat exhaust valves frequently in those engines, as racing pushed cylinder-head temperature high enough to experience some creep.

BY THE NUMBERS

10-25 HOURS THE TIME-BEFOREOVERHAUL LIFE OFTHE GERMAN JUMO 004 TURBOJET USED IN THE WWII ME 262

1,000-2,000 HOURS TBO OF THIRDGENERATION TURBOJETS SUCH AS THE P&WJ57 THAT POWERED MANY B-52s

50,000 HOURS NOW BEING TALKED ABOUT AS ATTAINABLE TBO OF FORESEEABLE FAN-JET ENGINES

This should be no surprise, as every time the power of air-cooled aircraft piston engines was raised by development, more fin area had to be added. In the case of Wright’s R1820 engine (which powered B-17S), cooling fin area per cylinder quadrupled over 10 years of development. Streetbikes are given fin area appropriate to their purpose, so racing, even with “stock” engines, can push head temperature above the range for which the engine was planned.

When large air-cooled engines powered the first Superbikes in the 1970s, it was normal practice in engine preparation to remove all casting flash from between cooling fins, especially in critical areas such as those giving access to the areas around spark plugs. This, by increasing airflow through fin spaces, offset some of the higher thermal load imposed by racing.

Liquid-cooled engines have thermostats, so within the limits imposed by a machine’s radiator size, as more power is used, the thermostat simply opens more to circulate more water through the radiator. When the limit of the existing radiator is reached, a bigger one—like the monsters seen in World Superbike—must be substituted. Familiar air-cooled motorcycle engines (that is, those lacking controllable blower cooling such as found on air-cooled Porsches) have no such thermostatic capability. This means two things happen. One is that as the machine moves faster in racing use, there will be some increase in cooling airflow through engine fins, but because the power required to overcome aero drag rises as the cube of speed, even increased cooling airflow falls behind the engine’s rising heat output. Because of this, the only way the hot parts can rid themselves of that heat is to become hotter.

I LEARNED THAT IT WAS ESSENTIAL TO RE-SEAT EXHAUST VALVES FREQUENTLY IN THOSE ENGINES, AS RACING PUSHED CYLINDERHEAD TEMPERATURE HIGH ENOUGH TO EXPERIENCE SOME CREEP.

Just be glad your air-cooled cylinder head is high-conductivity aluminum. Harley’s late racing director, Dick O’Brien, told me that at Daytona in 1970 the cylinder-head temperature of the first iron XR75OS ran as high as 900 degrees Fahrenheit. The aluminum heads of engines on WWII B-29 bombers were redlined at 520 degrees Fahrenheit but were driven as high as 640 degrees Fahrenheit by summer conditions in India. Cylinder head metallurgy today has not advanced drastically since 1944, so creep is still with us. And, just as in 1944, the most accurate predictor of future exhaust valve trouble is cylinder leak-down testing.