Perform rainflow cycle counting and then re-orders the cycles in the order in which the component experiences them. The operation of this option is identical to that described in Fatigue Preprocessing, 277.

Performs the actual crack growth analysis. This section describes this in detail.

The combined response time history can be output to a file. See Output Time Histories, 315.

The output response can be plotted. See Graphical Display of Time Histories, 318.

These utilities are identical to those described in Utilities, 324.

Exits the program.

The default will be the last jobname used in conjunction with a MSC Fatigue crack growth analysis. The jobname will pick up an existing jobname.tcy file which contains the local stress information. The jobname.tcy file is produced by the program FEFAT, the crack growth fatigue preprocessor. See Fatigue Preprocessing (Ch. 5).

This factor allows for scaling the stresses linearly to simulate the effect of a higher level of loading. The default answer is 1 (i.e., leave the stresses unscaled).

The mean offset of the stresses may be altered here to simulate the effect of a further mean load or residual stresses. Only positive offsets are allowed. The default answer is 0 (i.e., no mean offset will be added). Remember that crack growth is highly sensitive to mean stress since the stress is raised to a power of m in the calculation of crack increment where m is the Paris law slope.

The materials data describing the crack growth rate relationship are obtained from test coupons which are stresses in a plane strain condition. A correction may be applied to modify these data to simulate a plane stress condition. The default answer to this question is N, or NO (i.e., no plane stress correction will be applied).

If history effects are to be modeled, answer Yes to this question. Modelling the history effect means that the growth in one cycle is affected by the previous cycle history leading to retardations, accelerations, overshoots and other effects on growth rate and hence propagation life. It may be desirable to “turn off” the history effect to compare life predictions based on linear damage accumulation (no history effect) and those with history effects modeled.

If a results file is desired answer YES.

The PCRACK analysis generates a results file which will have the default name jobname.crg. An alternative name may be supplied in response to this question. If an output file already exists with the same name as the one supplied, overwrite permission will be requested.

A snapshot of the calculation parameters will be taken at regular intervals and stored in the output results file (jobname.crg). The user may select how often the snapshot is taken by defining the number of life units between snapshots. The default will cause the gap to be between 100 and 1000 cycles, converted into user life units. If the snapshots are too frequent, the crack growth results postprocessor (PCPOST) allows the user to decimate or compact the results file to save disk space.

The graphical status display shows the rate of crack advance using an a-N graph and various computational parameters. If the user requests not to have a graphical display, a tabular display will be produced indicating crack length, elapsed life in repeats of the loading history and elapsed life in user defined units.

The PCRACK graphical screen display will be updated at regular intervals. The update frequency should be chosen to reflect the speed of the computer on which the analysis is being carried out. An update interval of between 100 and 1000 cycles will be chosen using the default. Faster computers will require longer update intervals to prevent the computations becoming bound by the screen update rate.

The crack growth analysis may be terminated after a set number of user life units as specified here. If the number of life units is left undefined as for the default answer, the analysis will proceed until another failure criterion is satisfied.

The K solution is generated by the program PKSOL and is stored in a file with file type filename.ksn. The K solution relates crack size and compliance. For more detail, refer to K Solution Library (PKSOL), 508. The default K solution file name offered will be the last K solution used or generated.

The local geometry will now be defined. Note that the local geometry fields do not appear until the K solution file name is supplied and accepted. The length units will be reported to you as already defined in the K solution file. For example:

Units of length: inches

The user will then be asked to define the physical size of any additional notch not modeled using FEA and not accounted for in the compliance function.

If the crack under consideration is a surface crack, then a surface notch can be introduced into the global geometry from which the crack can grow. The crack will be influenced by the presence of the notch and its associated stress field. To correct the K solution for this, the local geometry of the notch has to be defined. The notch depth must be defined in the same length units as used for the crack definition when generating the K solution. Valid units will be either mm, m, milliinches or inches. If a notch is not present, (i.e., notch depth is zero), then the next two questions will not be asked and their defaults will be used.

This input is the notch root radius which defaults to 0.000039in if the Notch Depth above is zero. It must be defined in the same length units as used for the crack definition when generating the K solution.

This is the user-specified value of the root radius of a “sharp notch” (i.e., a real fatigue crack, which defaults to 0.000039in if the Notch Depth above is zero). The sharp crack root radius must be defined in the same length units as used for the crack definition when generating the K solution. Valid units are as for the previous question.

The default maximum crack size (n) is the component dimension in the cracking direction. This field enables you to specify a smaller value to take into account plastic collapse, yielding, etc. Note that if fracture toughness is exceeded at a crack size less than the minimum, then this condition will define the end of fatigue crack growth. All crack lengths MUST be defined in the same units as used for crack geometry when creating the K solution and you will be reminded of these units at the top of the input screen.

This is a critical value. It should be either:

1. The size of a discovered defect.

2. A postulated defect.

3. The NDT limit (i.e., the largest defect that cannot reliably be detected by nondestructive testing).

The initial crack size may be chosen in the range 0 to n where n is the maximum possible physical length of the crack based on the component geometry. A zero initial crack size may only be specified if a notch has been defined in the local geometry definition (see above). If no notch has been defined and you enter zero for the initial crack size, then PCRACK will compute the short crack parameter, lo, and substitute this for the initial crack size. See Cycle-by-Cycle Approach, 579, for details on lo.

This option allows users to select materials data from a user-defined database. The default is to search the standard database, which will exist in either the local or central directory.

The name of the user-defined database to use. This database may exist anywhere on the system, but the default is to look in the local directory.

The material name is the name used in the materials database under which are stored the parameters defining the crack growth properties of the material. The default name offered will be the last materials data set used. If the user wishes to specify another material, he may either type in a name that he knows exists or may use the List capability to select a material from a list.

The screen that appears when a materials listing is shown allows the user to view not only all the available materials but also to view the material parameters associated with each material. This eliminates the need to run PFMAT to view these parameters during an interactive run.

There may be crack growth materials data sets stored in the database for a range of different environments. The user may chose which environment to use in the analysis. If an environment is specified for which there are no data in the database, the parameters for AIR will be assumed and an environment definition will be requested (see Environment Definition below). If the user wishes to specify another environment, he/she may either type in a name that he knows exists or may use the List capability to select an environment from a list.

This option allows the user to edit the materials data for the selected environment.

This option gives a screen of information as shown in Figure 7‑35. The results are saved in a jobname.crg type file.

This gives the autoscaled final graphics plot ( Figure 7‑34 ).

This option gives the user the opportunity to interpolate the crack size and elapsed life data for different values of either the initial crack size or the final crack size or both. The defaults offered in the following example (Figure 7‑37) are the original initial and final crack sizes. It is possible to interpolate life beyond the crack size limits of the initial calculation. However, such interpolation is only approximate as the crack size-life relationship is both nonlinear and bounded by certain physical constraints.

As the title suggests, this loops the user back to screen 1 of this program and allows him to go through the whole process again.

This will bring up a submenu ( Figure 7‑38 ) which will allow for editing parameters associated with the initial job setup each of which has been discussed in detail previously.

This option allows for turning material checking on or off, setting a failure criterion, or doing an auto-overwrite.
The material parameters generally have certain valid ranges which the program checks for. If the user enters a value out of the valid range he will be told and requested to enter a valid number. He may see what the valid range is for a given parameter by using the Help on Question option from the help pull-down menu on the top bar when the appropriate field is active. If the user does not want to be bothered with this, he must set the material checking preference off ( Figure 7‑39 ).
The failure criterion option determines whether residual k is included in the calculation of k to exceed klc.
The auto-overwrite option sets a flag for this job. If the answer is Yes, then all output files are automatically deleted without conformation. If No, the program will request confirmation prior to deleting any output files.

After editing parameters, changing materials, etc., the user will need to select this option to see the effect of any of these changes. This reruns the analysis.

Allows for exiting the program and puts the user back either into the MSC Fatigue forms or back to the system prompt from which PCRACK was invoked.