A computational study of the influence of surface roughness on material strength

Abstract

In machine component stress analysis, it usually assumed that the geometry specified in CAD provides a fair representation of the geometry of the real component. While in particular circumstances, tolerance information, such as minimum thickness of a highly stressed region, might be taken into consideration, there is no standard practice for the representation of surface quality. It is known that surface roughness significantly influences fatigue life, but for this to be useful in the context of life prediction, there is a need to examine the nature of surface roughness and determine how best to characterise it. Non-smooth geometry can be represented in mathematics by fractals or other methods, but for a representation to have a practical value for a manufactured component, it is necessary to accept that there is a lower limit to surface profile measurement resolution. Resolution and mesh refinement also play a part in any computational analysis undertaken to assess surface profile effects: in the analyses presented, a nominal axi-symmetric geometry has been taken, with a finite non-smooth region on the boundary. Various surface roughness representations are modelled, and the significance of the characterized surface roughness type is investigated. It is shown that the applied load gives rise to a nominally uni-axial stress state of 90% of the yield, although surface roughness features have the effect of modifying the load path, and give rise to localized regions of plasticity near to the surface. The material of the test model is assumed to be elasto-plastic, and the development and evolution of plastic zones formed within the geometry are shown for multiple load cycles.

Synchrotron radiation microcomputed tomography for assessing internal cracks in cold spray repairs

“…4,6,9 However, as noted in MIL-STD-1530D, 10 it is virtually impossible to specify and analytically model all the types and sizes of initial defects that exist in aircraft structures. 4,6,9 However, as noted in MIL-STD-1530D, 10 it is virtually impossible to specify and analytically model all the types and sizes of initial defects that exist in aircraft structures.…”

“…The interaction between surface-breaking and nearsurface (subsurface) material discontinuities, ie, porosity due to lack of fusion or keyholing, and the rough surfaces that are associated with both AM replacement parts and cold spray repairs has the potential to complicate the analysis needed for certification. 4,6,9 However, as noted in MIL-STD-1530D, 10 it is virtually impossible to specify and analytically model all the types and sizes of initial defects that exist in aircraft structures. To overcome this difficulty, the concept of a surrogate crack size, also referred to as an equivalent initial defect size (EIDS), that accounts for all of these crack sizes has been introduced both for conventionally manufactured parts 10 and for AM replacement parts.…”

Combined effect of both surface finish and sub‐surface porosity on component strength under repeated load conditions

“…As the purpose of this study is to consider how this repeated loading might contribute to our understanding of fatigue life, in the analyses presented here multiple loading steps were defined, to give five fully reversed half cycles. Previous studies 77 indicate that, for higher numbers of cycles, the area of non‐zero PEEQ remains constant, but the values of PEEQ increase. This observation might be a consequence of the choice of plasticity model used.…”

“…Away from the region of interest, a much coarser mesh was sufficient, and geometry partitioning was employed to ensure a suitably smoothly graded mesh transition. This approach has been described and validated by the authors in more detail in previous work 71,77 . A basic principle in meshing regions where there is expected to be a high stress gradient is to ensure that the mesh is fine, and that the elements are of near equal size in such regions.…”

“…The surface roughness was defined randomly at 12.5 μm intervals, in a range of ±50 μm from the nominal surface, using the same method as described in Reference 77. In the current article, the profile is defined by a discrete set of points, through which a spline interpolation is fitted.…”