What is thermal fatigue?
Thermal fatigue is cracking caused by repeated stresses from temperature changes, such as hot and cold fluids mixing or a vessel heating and cooling. Hot fluid layered over cold in a pipe has the same effect. Each cycle may be harmless alone, but cracks start after many cycles, usually at welds and changes in shape.
By Jun Zhong, Editor·Researched and drafted by Beacon's AI newsroom·Checked by Jun Zhong, Editor·Last checked ·Next review
Why does it matter?
On 9 December 1987, a US nuclear power plant in Alabama, running at 33% power, found a leak of about 0.7 gallons per minute (about 2.6 L/min) from a crack that ran around and through the wall of a short section of emergency cooling piping 1. The US Nuclear Regulatory Commission (NRC) traced it to high-cycle thermal fatigue: relatively cold water leaking past a closed valve layered inside the hot pipe, and the temperature at the failed weld swung by as much as 70 °F (about 39 °C) every 2 to 20 minutes 1. The NRC noted that this pressure boundary had failed long before the end of its design life 1.
What causes thermal fatigue?
Fatigue needs a stress that rises and falls many times. It can come from mechanical sources such as pressure cycling, or from temperature differences that come and go 2. The cracking process is the same; only the source of the cycles differs. The US National Institute of Standards and Technology (NIST) lists these common sources of temperature cycling 2:
- Intermittent or periodic flow, where hot and cold fluid alternate. 2
- Frequent start-stop operation of the equipment or unit. 2
- Leaking valves, which let fluid at a different temperature into a line. 2
Two flow patterns recur in failure data. Mixing, where streams at different temperatures meet, and stratification, where hot and cold fluid form layers in a pipe 1,3. In its 2019 review, the piping failure database project (CODAP) of the Nuclear Energy Agency (NEA) of the Organisation for Economic Co-operation and Development (OECD) counted 254 pipe failures attributed to thermal fatigue, 80 of them (31%) after 1998 3.
Why is thermal fatigue easy to miss in design?
The loads are often not in the design basis. Because valves often leak, the NRC warned that an unrecognized and possibly unanalyzed condition could exist, and asked plants to review piping for temperature layering or swings that had not been evaluated in the piping design analysis 1.
Duty also changes over a plant's life. NIST notes that changes in production schedules, or rerouting of flow outside a vessel, can intensify cyclic stressing until a condition that was previously harmless becomes critical 2.
Test data can mislead as well. In tests on stainless clad material from coke drums, life under combined temperature and load cycling was shorter than life at a constant 480 °C, so estimates based on constant-temperature fatigue data were judged inaccurate 5.
Where does thermal fatigue occur in oil and gas plants?
- Delayed coker drums. Every cycle the drum is preheated by vapor from about 150 to 360 °C, filled with residue heated to about 482 °C for 14 to 18 hours, steam stripped as it cools from about 450 to 250 °C, then quenched with water, cut and reheated 5. The US Occupational Safety and Health Administration (OSHA) notes that the constant heating and cooling can make coke drum shells bulge and crack 4.
- Branch lines and dead-end sections connected to hot systems and closed off by valves that can leak 1.
- Mixing points, where streams at different temperatures join 3.
- Vessels and tanks with intermittent flow or frequent starts and stops 2.
How is thermal fatigue found and managed?
The NRC asked plants to take three steps 1:
- Screen. Review connected piping for sections that leaking valves could expose to temperature layering or swings. 1
- Inspect. Use non-destructive testing on the welds, the heat-affected zones beside them and other highly stressed spots such as abrupt changes in geometry. 1
- Control. Either redesign the piping to carry the combined loads, install temperature instruments with set limits to detect adverse temperature distributions, or monitor pressure upstream of block valves that might leak. 1
NIST adds that fatigue cracks tend to start in regions of high stress such as changes in geometry and at or near welds, which is where inspection looks 2. The NEA review stresses close cooperation between plant designer and owner, and between maintenance and operations staff, to control thermal fatigue risk 3.
Where integrity problems show up
| Where it occurs (unit or equipment) | Conditions that drive it | How it is usually found | In the Library |
|---|---|---|---|
| Branch lines on hot systems, closed by valves that can leak | Cold leakage layering in the pipe; temperature swings at welds 1 | Non-destructive testing of welds, heat-affected zones and shape changes 1 | What are the main non-destructive testing (NDT) methods in oil and gas? |
| Mixing points and lines with layered hot and cold flow | Mixing and stratification 1,3 | Temperature instruments with limits that flag adverse distributions 1 | What is a process data historian? |
| Delayed coker drums | Heating, filling, steam stripping and water quench every cycle 4,5 | Shell bulging and cracking; life checks need combined temperature and load data 4,5 | How are pressure vessels inspected? |
| Vessels and tanks with intermittent flow or frequent start-stop | Temperature cycling, made worse by later changes in production or flow routing 2 | Inspection at welds and changes in geometry, where fatigue cracks start 2 | What is management of change (MOC)? |
Sources: NRC Bulletin 88-08, 1988 1; NIST SP 780, 1990 2; NEA/CSNI/R(2019)13 3; OSHA Technical Manual, Section IV, Chapter 2 4; Chen and Xia, ICF13, 2013 5.
Common questions
- Is thermal fatigue the same as corrosion fatigue?
- No. Thermal fatigue is purely stress related: the cycles come from temperature differences. Corrosion fatigue needs a corrosive environment that speeds crack growth, such as hot water with dissolved oxygen for carbon steel. See [What is corrosion fatigue?](/library/corrosion-fatigue/).
- Can modest temperature swings cause it?
- Yes. In the 1987 case, swings of up to 70 °F (about 39 °C) repeated every few minutes were enough to crack a weld through the wall.
- Is thermal fatigue the same as mechanical fatigue?
- It is the same cracking process with a different source of stress: temperature differences rather than mechanical loads such as pressure cycling. See [What is mechanical and vibration fatigue?](/library/mechanical-vibration-fatigue/).
Beacon explains these concepts in its own words for education. It does not reproduce or replace the standards cited, and Winda is not affiliated with or endorsed by the organizations that publish them. Engineering decisions need the current edition of the standard and a qualified engineer.
Sources
- 1.NRC Bulletin No. 88-08: Thermal Stresses in Piping Connected to Reactor Coolant Systems · US Nuclear Regulatory Commission (NRC) ·
- 2.NIST Special Publication 780, Guidelines for Pressure Vessel Safety Assessment (S. Yukawa) · US National Institute of Standards and Technology (NIST)
- 3.NEA CODAP Project Topical Report: A Review of the Post-1998 Experience with Thermal Fatigue in Heavy Water and Light Water Reactor Piping Components, NEA/CSNI/R(2019)13 (report page) · OECD Nuclear Energy Agency (NEA), Committee on the Safety of Nuclear Installations
- 4.OSHA Technical Manual, Section IV, Chapter 2: Petroleum Refining Processes · US Occupational Safety and Health Administration (OSHA)
- 5.Thermal Mechanical Fatigue of Coke Drum Materials (J. Chen, Z. Xia, University of Alberta), 13th International Conference on Fracture · International Congress on Fracture (ICF13 proceedings, hosted by Gruppo Italiano Frattura)