Creeping Toward Failure

[Admin]

Kevin Cameron has been writing about motorcycles for nearly 50 years, first for <em>Cycle magazine</em> and, since 1992, for <em>Cycle World</em>. (Robert Martin/)

In the past I’ve written about how creep, the slow yielding of metals subjected to both high temperature and stress, results in exhaust-valve seat distortion as seen in motorcycle and aircraft engines. This in turn results in leakage, valve “overtemperaturing,” and perhaps failure.

Temperature is the measure of atomic activity, regardless of the state (gas, liquid, solid) of the subject of the measurement. Atoms in a metallic crystal lattice are held in place by electrical interatomic forces, but the higher their temperature, the greater the average energy of the atoms vibrating around their average positions. When provided with enough energy, they overcome the bonds holding them in their places, and the solid material transforms into a liquid.

Why Metal Moves

Metals are unique in that their electronic structure fills them with a “gas” of electrons. These same loose electrons give metals their high electrical and thermal conductivity. The electron cloud also allows metals to be malleable: Unlike most other materials, metals can be bent or formed into desired shapes without breaking. This is because interatomic bonds broken by externally applied force can reform nearby with new partners during deformation.

On the other hand, this same ability to reform interatomic bonds makes high-temperature creep possible. Even at temperatures far below the material’s melting point, applied stress can cause the metal’s slow yielding. Statistically, vibrating atoms oscillate around a mean position. But if enough extra energy happens along, a given atom may break loose from that position and take up a new position, joined to new partners, in a direction that tends to relieve the applied stress. This process, repeated a myriad of times, can allow exhaust-valve seats to gradually ovalize in service, or to move off center. It can also cause gradual elongation of the whirling blades in gas turbines, eventually resulting in their tips rubbing against the housing. Creep is also responsible for the very slow movement of the Earth’s continents.

Related: What Dyno Testing Can Tell Motorcyclists

The inadequate engine development of the World War II Boeing B-29 Superfortress resulted in excessive cylinder-head temperature for long periods in order to take off, climb, and then cruise for hours at 30,000 feet. Redline temp was 500 degrees Fahrenheit (270 degrees Celsius). At high altitude the air may be very cold, but its low density robs it of much of its cooling effect. The result was creep in the aluminum cylinder heads, and its harmful manifestation was valve-seat distortion and hot-gas leakage. The rough-’n’-ready diagnostic was to set a team of airmen to “pull through” a given engine’s four-blade prop while others listened for the “swish” of air past valves that could no longer seat and seal properly.

Leakdown Testing

In certain production-based motorcycle racing classes, particularly those whose engines were designed with no thought of competition use, the successful teams have been those who realized early that cylinder-head creep was causing steady loss of compression through valve-seat distortion. For every event they reseated the exhaust valves and directly measured their engines’ sealing quality by frequent use of static leakdown testing. (By 1945, the B-29 technicians also adopted such testing.)

For a static leakdown test, a subject cylinder’s piston is placed at TDC with all valves closed. A fixed pressure is then admitted to the cylinder—often through a fixture screwed into the spark plug hole—which is then disconnected from the pressure source. The pressure remaining after a chosen period is noted. When someone says, “Leakdown is 7 percent on number three,” it means that cylinder loses 7 percent of the test pressure in X time. Leakdown testers are commonly available from engine service-tool outlets.

Right from the very first days of the internal-combustion engine, creep of the exhaust valves themselves was also a critical problem, and the larger the valve, the more serious the problem. What engineers discovered was that valves, held against their seats by springs, gradually became longer as a result of 1) their very high operating temperature, and 2) the spring’s applied stress. This valve stretch, or creep, was easily measured by monitoring each valve’s lash or operating clearance.

Setting Valve Clearance

What is valve clearance? Because metals expand as they are heated, there must be clearance between the valve tappet and the cam lobe when the engine is cold. Were there none, once the engine has warmed up, thermal expansion would hold the valve slightly open. (In many auto and some motorcycle engines, automatic hydraulic clearance adjusters are built into each tappet so that valve clearance is maintained close to zero at all times.)

If, for example, exhaust valve clearance is set with 0.010 of an inch clearance when cold, and following some hours of operation the cold clearance measures as 0.006 of an inch, we can conclude that some combination of the following has taken place:

1. Valve-seat recession—the sealing surface between valve and seat wears, allowing the valve to move slightly in the direction of reduced clearance.<br/>
2. <a href="https://www.cycleworld.com/story/bikes/how-pneumatic-valve-systems-work/">Valve-stem stretch</a>—that some creep has occurred in the hottest areas of the valve, allowing it to become permanently a bit longer.<br/>

As the specific power of piston engines (horsepower per liter) increased steadily through the first half of the 20th century, there was a commensurate rise in exhaust-valve operating temperature. Supercharging, adopted in large aircraft engines and racing engines between 1919–1939, further raised valve temperature. Improvements in valve materials alone failed to keep up, so exhaust-valve creep became an obstacle to both high power (for takeoff or combat) and to long operation. During WWII, Consolidated B-24 Liberators flew anti-submarine patrols as long as 20 hours.

Sodium-Filled Valves

The solution was to use a partial liquid filling inside the exhaust valve stem. As the valve opened and closed, by its own inertia the liquid filling tended to stay somewhere in the middle, alternately picking up heat from the critical hot region where the stem flares to join the valve head, then transmitting that heat to the cooler end of the valve. This, by reducing temperature in the critical area, could greatly slow creep. Sodium metal proved best for this filling, with a melting point of 208 degrees Fahrenheit or 98 degrees Celsius.

Related: Valve-Seat Rings on the Loose!

This is why early high-power engines were limited in the length of time they could safely deliver peak power. A favorite example, which puzzled me for years, was the Rolls-Royce R V-12 engine, developed for use in the Schneider Cup air races. In its case, operation was limited to 30 minutes. Even much later, in commercial aviation during the 1950s, the use of takeoff power was limited to a maximum of five minutes.

Titanium Valves

Similarly, when motorcycle racing engines started exploring titanium as a valve material, it was initially used only for intake valves, with exhausts still made of heat-resisting stainless steels. The next step was to add titanium exhaust valves, but only for sprint racing, and to continue using stainless exhaust valves for the longer endurance events. As best practice was developed over time, some production sportbike engines have adopted titanium exhaust valves.

In large aircraft engines, the severe conditions arose from supercharging, long periods at high power, the limited effectiveness of air-cooling, and from the exhaust valves’ large size (i.e., long heat paths), as big as 3-5/16 inches (84mm) in diameter.

Modern 1,000cc sportbikes have exhaust valves with heads about equal in diameter to that quarter. Their heat path to the seat is a lot shorter than that big radial aircraft engine's valve! (Mark Lindemann/)

Even much smaller air-cooled motorcycle engines continued to have problems with exhaust valves into the 1950s. The exhaust valve of BSA’s classic 500cc Gold Star was upgraded to the then-new gas-turbine alloy Nimonic 80A. Norton sought to improve the rate at which sodium in its 500 Manx exhaust valve could pull heat out of the hot end and move it up the stem, from there to be transmitted to the cooler valve guide, which in turn is pressed into the cylinder head, the exterior of which is covered with cooling fins. In 1953 they circulated oil through a spiral passage around the exhaust-valve guide and ran it through a small frame-mounted cooler.

Related: The Coming Crunch In Big Twin Engine Cooling

All this action just to pull valve temperature down enough that the process of high-temperature creep was sufficiently slowed to permit the desired performance.

We have it easy today, when most motorbike engines are liquid-cooled and their exhaust valves are so small that the combination of short heat path and liquid-cooled cylinder-head material make them all but immune to creep failure.

View the full article