Possibilities and Limits
Certification Procedures: Static / Endurance Limit / Fatigue Life / Crack Growth
There are several steps in the development process of a component. In a first step you have to design the component for sufficient safety to keep the maximum of static loading. A safety factor is used to take uncertainties into account. This static proof is the first step and its successful certification is necessary for all following procedures.
If dynamic loadings are acting on the component it also has to be proven that no damage occurs under dynamic conditions. In the very early stage of development usually only limited component information is available, e.g. its material properties, way of production, loading.
The Endurance Limit Proof can possibly help in this situation because it is possible to safely show that in a worst case scenario the stress is below the endurance limit. This result is often sufficient and more extensive fatigue strength tests are no longer necessary.
Later in the development process – when the component exists – much more information about the component is available and it is often possible to take measurements during operation. If the endurance certification fails or the safety against the endurance limit is not sufficient then a fatigue life calculation must be done.
For special components it may be necessary to know the crack growth behaviour e.g. to define inspection intervals during operation.
For all the above mentioned scenarios winLIFE supports modules for analysis as shown in the following table:
|
Goal |
Result |
Procedure |
winLIFE Module |
Remarks |
|
Static analysis for welded and non-welded components
|
Safety against maximum loads |
Utilization ratio of the static limits. |
winLIFE Quick Check |
According FKM guideline |
|
Fatigue analysis for welded and non-welded components
|
Safety against damage for a simple given load (spectrum) |
Utilization ratio of the fatigue limits. |
winLIFE Quick Check |
According FKM guideline |
|
Endurance Limit Certification also for non-proportional cases |
Safety against endurance limit |
Worst-Case analysis |
winLIFE Quick Check |
Not according FKM |
|
Fatigue Life prediction |
Life (in hours, km, number of repetitions, etc.) |
Analysis of real loading, super positioning of loading
|
winLIFE BASIC
|
Uniaxial / Biaxial Proportional |
|
winLIFE MULTIAXIAL (Extension for winLIFE BASIC) |
Rotation of principal stresses
|
|||
|
winLIFE GEARWHEEL & BEARING (Extension for winLIFE BASIC) |
Tooth foot (break) / tooth flange (Pitting) |
|||
|
Crack Growth |
Crack growth after first crack appears until break
|
|
winLIFE CRACK (Extension for winLIFE BASIC) |
|
|
Random Analysis using PSD |
Life (in hours, km, number of repetitions, etc.) |
For a given PSD the damage for each node is calculated |
winLIFE RANDOM |
If the exciting frequency is higher than 1/3 of the smallest eigenfrequency |
Static Certification according to FKM
To perform the static certification the static material data such as RM, RE and allowable strain are used. The analysis takes into account the ductility of the material and gives a good estimation of static safety.
Fatigue certification according to FKM
The fatigue certification is done for one point of the structure. The user has to select this point e.g. coming from FEA. The characteristic properties for this point of the component and the material properties are used to create an S-N curve. If there is more than one component of the stress tensor for each stress a utilization ratio is calculated and finally a weighted addition to get a total result.
The loading may be a load spectrum for up to 6 steps.
Endurance Limit Certification
The Endurance Limit Certification can help to give an estimation on how the structure is loaded compared to the endurance limit. If it is possible to safely show that in a worst-cast-scenario the stress is below the endurance limit, then this result is often sufficient and more extensive fatigue strength tests are not necessary.
The endurance limit certification cannot give you the fatigue life, but a prediction as to whether the worst-case-scenario is below the endurance limit and if so, by how much. If the endurance limit certification is not fulfilled, then it will be necessary to carry out much more detailed fatigue life tests. For these, you will need, in particular, the loads and their time relation.
Fatigue Life Calculation
Fatigue life calculations give good relative information in the early development stages. Different designs can be calculated easily and predictions of fatigue life can then be carried out. As a result the development process of dynamically loaded components can be shortened considerably.
Numerical fatigue life predictions are currently not accurate enough to replace testing necessary to prove fatigue strength because the absolute number of cycles is not accurate enough. Because of safety reasons it will be necessary to test a component under service load in addition to calculating the fatigue life.
A lot of fatigue calculations and test results were examined to give an idea about the differences between test and calculation. It was found, that the results are in the range of 0,1 to 10 compared to the rality. This means that the calculated fatigue life may differ with a factor 10 or 1/10!! A fatigue life prediction only based on fatigue calculation will not be precise enough in most cases.
In the case you have test results from your component in reality (a statistally sufficient number) you can define a correction factor which scales the calculation results to the reality. This factor doesn´t change if you have “similar” problems.
Crack Growth
Components made from steel materials as these are used in civil engineering or automotive industry show a short cycle number (time) between crack initiation and total failure. This is the reason why the design is only done until crack initiation and the short period of crack growth is not of interest. The use of the component ends if a crack occurs.
On the other hand, in areas such as airplane or airspace industry where aluminium alloys are used, the cycle number (time) from crack initiation until failure is remarkably high. In this case the use of the component in the period of crack growth is possible and efficient and so it is done. But crack growth reduces the supporting area of the component and there is a critical crack length which leads to total failure.
The calculation of crack growth helps to design the component without increase of the critical crack length.
Random Analysis
In many areas of engineering stochastic loadings exist and these are described by the PSD (Power spectral density). In planes, ships or land vehicles this way of desribing a load is common. Using FEA you can calculate the the behaviour of components excited by a PSD.
The dynamic structural analysis leads to peak stresses in that range where the exciting frequence is near to the normal frequencies. This is a great advantage because related to the static analysis or the static superposition whrer you do not get any influence of frequency.