The temperature change in the operation of the pipeline may cause the thermal expansion and cold contraction of the pipeline and equipment, which also causes the expansion and deformation of the bellows. In some cases, the metal bellows will also expand and deform due to mechanical displacement. The stress caused by these displacement changes is often alternating stress. Metal bellows may cause fatigue failure under alternating stress. For metal bellows, due to the large displacement change, the range of alternating stress caused by it is also large, which is easy to cause fatigue damage, so the fatigue of bellows has become a problem that must be seriously considered in design calculation. There are generally two types of formulas or curves for the design of metal bellows fatigue: one is for empirical theoretical formulas, such as the Langer formula derived on the basis of Manson-Coffin formula. The other type is the formula or curve based on experimental data, such as the calculation formula and curve in the EJMA(American Expansion Joint Manufacturers Association standard) standard. According to the actual application situation, the calculation formula of EJMA is close to the actual situation, but various design formulas or diagrams have certain applicable conditions. At the same time, there are many factors affecting the fatigue life of the metal bellows, and the data are scattered. Therefore, in order to actually assess the fatigue performance of the metal bellows or to test that the bellows can withstand a predetermined number of deformation cycles, it is necessary to carry out fatigue tests on the metal bellows. Fatigue life is also an important index to evaluate the quality of metal bellows.

1) Specimen

Metal bellows fatigue test specimen, generally should be the actual product. As a test piece of metal bellows products should be representative of qualified products. The wave number of the specimen should not be less than 3, the waveform of the bellows and the manufacturing process should meet the relevant standards, and the various parameters of the bellows should be fully grasped before the test.

2) Determination of fatigue test conditions

The fatigue test of metal bellows should be carried out on a dedicated fatigue test machine. Currently, the commonly used fatigue test machine has two categories: hydraulic or mechanical.

For metal bellows whose design temperature is at the material creep temperature, the test can be carried out at room temperature.

In the fatigue test of metal bellows, the bellows can be normal pressure, or can be used to change the pressure. The latter is closer to the actual use of metal bellows. The cyclic displacement can be a symmetric displacement cycle of tension and pressure, or a displacement cycle of axial compression from a free length to a specified value. When determining the cycle test rate, the displacement should be evenly distributed among the waves, generally not more than 30 times per minute is appropriate.

In the fatigue test of metal bellows, when there is a penetrating crack in the bellows, it is considered that fatigue failure has occurred. At this time, the cycle life measured is the actual fatigue failure life of the metal bellows (times).

In the fatigue test, the deformation of each wave of the metal bellows may be inconsistent, so it is necessary to observe the change of each wave pitch. The fatigue failure life of metal bellows can be calculated according to the actual measured value of cyclic displacement of each wave distance and the actual size of metal bellows. At the same time, in order to facilitate analysis, it is also calculated according to the nominal size on the drawing. Finally list for comparison.