Engineering task and calculation objective
The EK17 module calculates the fatigue life of pressure vessels using the simplified method of DIN EN 13445-3 clause 17. If you want to calculate the allowable number of load cycles of a vessel subjected to cyclic pressure fluctuations, you get a conservative estimate here based on stress ranges and the fatigue curves (S-N curves) of the standard. The method distinguishes between welded components, unwelded components and bolted connections, since these follow different fatigue curves.
In practice, this fatigue assessment is required whenever a vessel is not predominantly statically loaded, for example in batch processes, pressure-swing adsorbers, autoclaves or plants that are frequently started up and shut down. Even the initial check of whether a fatigue assessment is needed at all (the cycle-number threshold in EN 13445-3 clause 5.4.2) relies on this simplified method. Influencing factors such as ovality and peaking (angular misalignment) at weld seams, as well as the weld execution, e.g. a flush ground butt weld, enter the result directly via the fatigue class and correction factors.
Because the method deliberately makes conservative assumptions, it often yields considerably lower allowable cycle numbers than the detailed fatigue assessment of clause 18 (module EN18). If the simplified fatigue life is sufficient, the verification is complete with minimal effort; otherwise it pays to switch to the detailed calculation with structural stresses.
Standard and calculation basis: DIN EN 13445-3/17: 2021-12
Calculation workflow
- Record load cycles and pressure ranges: First, the cyclic loads are defined as pressure ranges with associated numbers of cycles. Several different cycle types, such as full-pressure cycles and smaller operating fluctuations, are recorded separately and later combined via damage accumulation.
- Determine the pseudo-elastic stress range: The governing stress range of the component is derived from the pressure range. In the simplified method this is done conservatively via the nominal design stress and stress factors, without requiring a detailed structural stress analysis.
- Classify the component: Welded components are assigned to a fatigue class that depends on the weld detail and the extent of testing. Deviations such as ovality or peaking (angular misalignment) at longitudinal seams increase the local bending stress and are accounted for via correction factors; a flush ground butt weld allows a more favourable classification.
- Apply correction factors: The standard provides corrections for wall thickness, temperature and, for unwelded components, for the mean stress effect. These factors reduce the endurable stress range compared with the reference S-N curves.
- Read off the allowable number of cycles: Using the corrected stress range, the allowable number of load cycles is determined from the governing fatigue curve and compared with the required number of cycles.
- Accumulate the damage: With several cycle types, the total damage is summed using linear damage accumulation (Miner's rule). The verification is satisfied if the damage sum remains less than or equal to 1.
Input quantities
| Quantity | Symbol | Unit |
|---|---|---|
| Material | Werk | – |
| Strength value | K | N/mm² |
| Safety factor | S | - |
| Nominal wall thickness (order wall thickness) | en | mm |
| Const in equations of calculation fatigue | C1 | N/mm² |
| Const in equations of calculation fatigue | C2 | N/mm² |
| Const in calculation fatigue curves of | C3 | N/mm² |
| Fatigue class from Table 17-3 | C | – |
| Theoretical notch factor | Kt | - |
| Number of pressure fluctuation widths the z | k | - |
| Actual number of stress cycles | nk | - |
| Transition radius | r | mm |
| Min. temperature during one cycle | tmin | °C |
| Max. Temperature during one cycle | tmax | °C |
| Calculated nominal stress | f | N/mm² |
| Stress factor | η | - |
| Maximum allowable pressure of the vessel or component | Pall | MPa(p) |
| Thickness of thinner wall (0 if not present or same wall thickness) | e2 | mm |
| Desired number of load cycles | N | - |
| Flush ground butt weld | weld | – |
| Parameter z according to test group | z | – |
| Check of allowable pressure cycles according to formula 17.5-2? | UEP | – |
| Stress range calculated according to Annex B or C of EN 13445-3? | UEC | – |
| Stress range calculated according to Annex B or C of EN 13445-3 | ΔσVG | N/mm² |
Calculated results
| Quantity | Symbol | Unit |
|---|---|---|
| Correction factor for considering wall thickness | ce | - |
| Correction factor for considering temperature | ct | - |
| Effective notch factor | Kf | - |
| Allowable number of load cycles | Nall | - |
| Assumed mean temperature during one cycle | t* | °C |
| Existing pressure fluctuation width | ΔP | MPa(p) |
| Fictitious stress range | Δσ* | N/mm² |
| Allowable pseudoelastic stress range | ΔσR | N/mm² |
| Fatigue strength at constant stress range | ΔσD | N/mm² |
| Allowable number of load cycles | Nall | - |
| Allowable pseudoelastic stress range | ΔσR | N/mm² |
| Fictitious stress range | Δσ* | N/mm² |
| Lower limit value of the stress range | Δσcut | N/mm² |
| Allowable pressure fluctuation width | Δpall | MPa(p) |
| Reference wall thickness | eb | mm |
| Pseudoelastic stress range | Δσ | N/mm² |
| Correction factor ke (18.8-1) | ke | – |
| Corrected pseudoelastic stress vibration amplitude | Δσcorr | N/mm² |
| Allowable number of full pressure cycles listed in 17.5.4.2 | Neq | – |
| Degree of fatigue | DF | -- |
| Condition | Condition | – |
| Allowable number of load cycles at given pressure listed in 17.5.4.2 | Nall' | – |
Calculation options
Flush ground butt weld
Yes · No
Check of allowable pressure cycles according to formula 17.5-2?
No · Yes
Stress range calculated according to Annex B or C of EN 13445-3?
No · Yes
Detailed description from table 17-2
1 · 2 · 3 · 4 · 5 · 6 · 7 · 8 · 9 · 10 · 11 · 12
Detailed description from table 17-2
16 · 20 · 35 · 40 · manual Input
Progression of the amplitude of the pressure variation
constant · variable (part of loading collective)
Brim present?
No · Yes
Thermal gradient type
Linear gradient through thickness or Linear gradient along surface direction in shell · Linear gradient along surface direction in flat end · Thermal shock
Frequently asked questions
When is the simplified method of clause 17 sufficient, and when do I need clause 18?
The simplified method is suitable as a conservative first assessment for predominantly pressure-induced cyclic loading. It delivers quick results but does not exploit the reserves of the component. If the verification narrowly fails, high cycle numbers are required, or temperature cycles and external loads dominate, the detailed assessment of clause 18 with structural stresses (module EN18) is the right choice.
Can I use the module to assess cyclic loads other than pressure cycles?
Yes, with restrictions. The standard permits converting cyclic loads from temperature changes or varying external loads into equivalent stress ranges and evaluating them with the same set of curves. The conservative assumptions of the simplified method and the limitations of the standard's scope must be observed.
Why do ovality and peaking of the weld seam degrade the result so strongly?
Ovality and misalignment generate additional local bending stresses in the seam under internal pressure, superimposed on the membrane stress. Since fatigue damage grows with a power of the stress range, even a moderate increase in local stress leads to a markedly reduced fatigue life. The standard captures this via stress magnification factors calculated from the measured shape deviations.
What is gained by grinding the butt weld flush?
A butt weld ground flush with the plate and inspected eliminates the notch effect of the weld toe and may be assigned to a higher fatigue class. This considerably increases the endurable stress range and thus the allowable number of load cycles. The prerequisite is that grinding covers the full surface and is backed up by non-destructive testing.