Engineering Failures Driven by Heat, Time, and Instability
(Level 4)
Some failure modes only appear under specific conditions: high temperature, compressive loading, or thermal cycling. Level 4 introduces failure mechanisms that are often overlooked in basic design but critical in real world applications.
At this stage, the focus is on understanding how time, temperature, and geometry create unique failure risks that require specialized analysis.
What You'll Learn
Creep: Time Dependent Deformation at Elevated Temperature
It starts with three simple words: heat, time, and creep. In technical terms, creep is the slow, irreversible deformation of a material at high temperature and under constant stress. If a part is subjected to creep, it slowly starts to stretch—inch by inch, over hours, days, or even years. Creep doesn't work like conventional plastic deformation, which can be determined by a single yield strength. Creep takes place gradually, steadily, over time, leaving no clear warning signs—until it is too late. For months, your part will seem to behave perfectly. Then, without warning, clearance will be lost, components will have to be realigned and severe interference will occur.
What is creep in materials? Creep occurs in materials when they are subjected to stress at elevated temperature. As a rule of thumb creep is significant when the absolute temperature is greater than about 0.4 times the melting point. This translates to around 400-500°C for steel. For Aluminium the creep temperature can be quite low (room temperature for some high stiffness alloys), whilst for some high temperature ceramics creep is only a problem when the temperature is raised even higher. Increased stress causes a greater creep rate, and time causes creep more than anything else. Thus an engine turbine blade operating at 900°C may be below the allowable static stress level, but after 10,000 hours will have crept so far that it will no longer meet the design specification or even suffer a catastrophic failure.
Creep is of particular importance in any high-temperature application, from turbines to industrial boilers to nuclear reactors. It directly affects engine cooling systems, exhausts, and can have a large impact on the overall service life of a part or assembly. Creep occurs over a long period of time, slowly causing a part to deform, loose shape, or eventually fail. Most creep failures do not happen immediately, they happen over a long, grinding period of time. This makes creep failures especially hazardous, as there is generally enough time for warning signs to appear.
The Three Stages of Creep and How Rupture Develops
Creep deformation typically develops in a well defined sequence of three stages. Initially, a short period of rapid deformation occurs or Primary creep. The material work-hardens, as dislocation movement leads to accumulation of dislocations, as a result the creep rate slowly decreases. This stage is very short, of the order of hours to days. The deformation then enters into a secondary or steady-state creep stage. The creep rate remains steady as work hardening and thermal recovery are in balance and material deforms at a constant rate. This is the longest stage accounting for majority of the component's life.
Eventually every alloy enters Tertiary creep (Stage 3), or tertiary creep, where deformation accelerates rapidly, leading to void formation, necking, cracking and eventual failure. The time to failure can be as short as minutes in some high-temperature situations and as long as years in others. For design, the goal is to keep parts in a state of steady-state creep (Stage 2 or secondary creep) and avoid entering the more critical Stage 3 creep. By controlling stress and temperature to keep the creep rate within reasonable limits, secondary creep rates can be maintained to ensure long-life design.
Creep design takes into account the secondary stage creep rate, which means you should design against that specific rate. Determine how much your part will deform over time (let's say 10 years or 50,000 hours) and design the part so it will be within tolerance before it reaches the tertiary stage of creep—the point where your part fails.
Buckling: Geometry-Driven Structural Instability
Most people are tripped up by buckling because they don't understand what buckling is. Buckling is a form of geometric instability where a long slender column is compressed beyond a certain point. Stress levels are completely safe for the material. It remains elastic, yet it can suddenly deflect excessively and buckle. The material hasn't failed, it hasn't fractured, it hasn't broken. It's simply gone geometrically unstable.
This is geometry dependent. Long, thin bars will fail by buckling at loads that are much less than they can handle in tension or in short compression. The critical buckling load is given by Euler's formula and follows the equation: Fcr = (n * pi^2 * E * I) / (L^2) where Fcr is the critical buckling load, n is a constant which depends on the end conditions of the bar, E is the modulus of elasticity of the bar, I is the moment of inertia of the bar about its axis, and L is the length of the bar. This equation shows that the critical buckling load is independent of the material strength. Thus, for example, a very long, thin steel column may have a critical buckling load much less than its yield strength.
A 5m long steel column with a square section of 200mm might buckle under as little as 10kN applied in compression. This is well below the expected yield stress of 1000kN/(200mm)². Doubling the section area will increase the buckling resistance by at most two times. Halving the length will increase the buckling resistance by four times. Geometry is the overwhelming factor. So, while increasing the section area will help, a simple static stress analysis is not sufficient to establish the design safety of a compression member. A separate analysis for buckling is required.
Thermal Stress From Constrained Expansion and Cycling
When temperature changes, most materials expand or contract in predictable ways. When embedded in a structure, however, the material's expansion or contraction is typically constrained so that the component grows or shrinks in size with temperature, but the rest of the structure does not. This restrained expansion or contraction directly results in thermal stress, which develops regardless of the external load or applied stress, even when the stress remains within commonly accepted limits. Railroad tracks buckling in hot summer weather, cracking and leaking pipelines splitting apart in cold winter weather, and bolted joints losing their clamp load after being cycled through hundreds of degrees in temperature—each of these problems is caused by thermal stress.
There are three general cases of thermal stress. One is called "restrained expansion." In this case, the part is fixed in place by bolting or welding. When heated, internal compressive stress builds up in the part until something yields or breaks loose. A second case is called "temperature gradients." In this case, different parts of the component are heated to different temperatures so that different parts attempt to expand to different degrees. Since the whole part is made of the same material, stress develops because some regions attempt to expand more than others. The surface of an aluminum casting for example might be at 200°F while the center is at 180°F. The surface wants to expand more than the center, and the stress builds up until cracks appear. The third case of thermal stress is when different materials are attached rigidly to each other. When heated, one material attempts to expand to a greater degree than the other material. In the case of aluminum bonded to steel for example, the aluminum attempts to expand to a length about two times that of the steel. This results in stresses that cause cracks or delamination.
Thermal cycling places similar stresses on a material as do rapid changes in applied load leading to mechanical fatigue. Repeated heating and cooling can cause fatigue failure similar to that observed in other load controlled fatigues. Furthermore, when thermal cycling is combined with mechanical stresses, the material's lifetime is still further decreased. At first glance, the failure of exhaust manifolds appears to be entirely mechanical i.e. a case of over-torque on dropping the car to the ground after liftiing for servicing. However, closer inspection reveals that cracking is due to cold starts on a hot engine. Substrates to electronics also frequently fail by power cycling, i.e. switching on and off the AC power. This also is a case of fatigue failure due to thermal cycling. The pattern of failure is similar to that observed in other fatigue loading schemes and is characteristically similar to that of any other fatigue failure due to causes of either a mechanical or thermal nature. The failure is due to theraml fatigue, a form of fatigue failure similar to mechanical fatigue but caused by temperature changes rather than loading.
Thermal Shock From Rapid Temperature Change
The term Thermal Shock is used to describe the damage that occurs to materials when there is a large change in their environment. You probably have seen what happens when you pour cold water into a hot dinner plate made of glass, and it shatters into pieces. This occurs due to the rapid change in temperature that causes the large stress gradients through the thickness of the material. Because of the instant change in temperature, the surface tries to expand or contract instantly, while the interior goes through a much slower process of expansion or contraction. Thus, even without any external loading, the internal forces due to the differential strain can reach tremendous levels and cause crack propagation and material failure.
Brittle materials like ceramics and glass crack instantaneously under thermal shock as they are unable to undergo strain. Ductile materials like many cast irons can undergo some level of strain but suffer loss of life due to repeated thermal shock in operations that involve multiple thermal cycles like a ramp up from cold start to full power in a gas turbine engine, or rapid variation in flow rate through heat exchangers. Turbine blade airfoils go from cold start temperatures to operational temperatures in a very short period of time, while the engine block undergoes a rapid heating due to the spark plug and gas combustion in the cylinders during start. Materials like these are subject to thermal fatigue where multiple cycles of thermal shock do not visibly effect the component but each cycle does accumulate some damage to the material.
The problem gets worse when the rate of temperature change increases. This is because slower cooling or heating allows the temperature to more evenly distribute throughout the cross-section. Rapid quenching or heating, on the other hand, puts the maximum stress on a part. Preheating mold-release trays, glass cookware's limits on the difference between successive temperatures, turbine startup ramps – none of these exist because sudden changes in temperature are innocuous. In fact, design that neglects thermal-shock resistance is likely to fail either immediately or in the not-so-distant future.
Design Strategies for Creep, Buckling, and Thermal Failure
Each failure mechanism has a different battle strategy. Creep - use high temperature alloys (Nickel based superalloys for turbines, ceramics for high temperature wear resistance). If these options are not available, reduce operating temperature or design for operating stress below yield point (design for some deformation over the life of the part - not for zero creep strain). Calculate the total strain after 50,000 hours and ensure that appropriate clearances have been incorporated into design.
For buckling: geometry is everything. For most structures, the simplest way to increase the buckling strength is to reduce the slenderness ratio – this can be achieved by reducing the length of the member, or by increasing the radius of gyration (e.g., by using a heavier wall thickness for the same outer dimensions). In some cases additional lateral bracing (e.g., stiffeners) may also improve the stability. Another common method to improve the buckling strength is to use cross sections with a higher moment of inertia (I) for the same weight, such as I-beams instead of channels, or hollow tubes instead of solid rod profiles. The high buckling strength of the tube comes from the fact that the structural material is located far away from the neutral axis, and thus is much stiffer than the solid rod made of the same material with the same weight. That's why steel mountain and racing bike frames have tubular profiles, not solid rod profiles.
When performing assembly which is subject to subsequent thermal cycling it is essential to allow for thermal expansion. In pipeline and bridge constructions this is achieved through the use of expansion joints. In machinery, flexible couplings are generally used to allow for thermal movement between pipework and rotating shafts. When bonding two different materials, it is desirable to try and achieve close thermal expansion coefficients, or alternatively use a compliant adhesive which can allow for the differential movement of the substrate materials. The rates of heating and cooling should also be controlled to limit the effects of thermal shock – rather than suddenly ramping up temperature, for example, use a slow and gentle build-up. Incorporation of thermal barriers or insulation can also be of use in protecting specific components from adverse temperature effects. To introduce rigid constraints on an item which is intended to expand, then subsequently heat it up, seems dumb, and invites cracking or plasticisation.
Why Time, Temperature, and Geometry Control Failure
High fatigue stress, Creep, buckling and high temperature, thermal induced failure are usually not addressed in typical design and released to production due to lack of analysis tools and capability to perform such detailed static FEA. As a result, after six months of actual use, some components loose tolerances, thin supports buckle or welds crack due to thermal cycling. This type of failures are catastrophic and costly because they occur after product is released to production.
This level of understanding of the common failure modes teaches you when to worry in each design mode. High temperature? Check for creep. Slender compression members? Calculate the buckling load. Thermal cycling or rapid temperature changes? Don't ignore the effect of thermal stress and the possibility of fatigue due to cycle of high and low temperatures. In designing to accommodate thermal expansion, you must also impose limits on net heating rates. And it may be that buckling is not the failure mode of primary concern. Buckling, creep, and fatigue due to temperature effects are common failure modes in power plants, aircraft, bridges, towers, rocket components, and many other structures.
After creep, buckling, and thermal stress, it's time for the last level: wear, corrosion, and reliability engineering. This is the realm of degradation over time, and designing against environmental loading to predict when your parts will fail so you can replace them before that happens. Most failures aren't sudden snap-load failures. They're slow failures due to slow degradation which you may not have thought about.