Learning Forum

[Duncan Gibbons]

Hi all,

I am trying to run a fatigue crack growth analysis. I am wanting to plot the crack extension against the number of cycles and determine after how many cycles the part fractures (when residual strength is below a set value). I am getting an error "For Cgrow Id 1 and Crack Id 1, the computed number of cycles is less than 1. Under this loading condition, static crack growth is expected. Please check your model, boundary conditions and fatigue parameters.  The crack growth solve is stopped. " when running a fatigue SMART crack growth analysis. When I run a static crack growth analysis on the same part, same loads & constraints, same crack dimensions etc. the results indicate that the crack wont grow and the SIF is well below the critical fracture toughness (by a magnitude of 10). Additionally, the maximum stresses on the part for the same load case are 174 MPa, well below the yield strength of the alloy of 850 MPa.

Does this mean that I have incorrectly input the Paris law constants or is there something else I am missing? I am using reference units of m, N. My Paris law constants are C = 9e-11 and m = 2.8 for a titanium alloy.

[danielshaw]

The message indicates that for current cylic loading and the current model parameters (crack size, SIFs, Paris Law constants, etc.) the crack will not grow.  If you want to propagate the crack, you need to increase the SIFs (probably by increasing the stress).

[John Macgregor]

 

It may be an issue with your Paris law units. The value of C depends on the units used to calculate the stress intensity factor range and da/dN for the Paris law curve, a C value of e-11 suggests that MPa sqrt m and m/cycle may have been used. For reference units of (metre , N) the SIF range should be in Pa sqrt m. An approximate conversion can be made by dividing your C value by (1×10^6)^m, which for your units would give a C value of 1.4264e-27, the value for m will remain the same. However, its important to double check what units were used in the derivation of your Paris law constants, as they may have used another unit system. 

 

[Duncan Gibbons]

Hi John, I suspected my units could be off. I used the units of "MPa" and "m" when calculating the Paris constants I am using. Many thanks for this suggestion.