Page 355 - Marine Structural Design
P. 355

Chapter I7 Fatigue Capacity                                           33 1

                 In some of the design codes, there is a cut-off limit, and low fatigue damage is assumed when
                 the stress range is below the cut-off limit.
                 For the sake of consistency, discussions of the fatigue criteria in this chapter will be mainly
                 based  on NORSOK (NTS, 1998). However, readers  are recommended to  refer  the  codes
                 relevant to their projects such as IIW (Hobbacher, A. (1 996), Eurocode 3 (1992), IACS (1 999),
                 ABS (1 992) and DNV (2000), among of others.

                 17.1.2  Effect of Plate Thickness
                 The thickness effect is due to the local geometry of the weld toe in relation to the thickness of
                 the adjoining plates and to the stress gradient over the thickness. It may be accounted for by:

                     logN=logK-mlog  S  -                                            (17.3)
                                      [ (tY]

                 where, tref   = Reference thickness which in some design codes is 32 mm  and 25 mm  for
                            tubular joints and other types of welded connections respectively  (NTS, 1998).
                      t     = Thickness through which a crack will most likely grow.
                      k     = Thickness exponent on fatigue strength in the range 0.00 to 0.25 depending
                            on the  code employed, the S-N curves selected etc (NTS,  1998).

                 In other words, the thickness effect may be accounted for by multiplying a factor of (r / t Rf )" to
                 the stress range. In HSE (1995), the value of k and reference thickness I,~, is 0.25 and 22 mm,
                 respectively. In general, the thickness correction to the design equation for the S-N curve is
                 required when the plate thickness is thicker than the reference thickness. To some extent, the
                 thickness correction also accounts for the size of the weld and its attachments. However, it
                 does not  account for the weld  length or the length of component different from the tested
                 component.
                 17.1.3  Effect of Seawater and Corrosion Protection
                 In  Figure 17.2, three types of S-N curve are compared for the tubular T S-N curves in  air,
                 seawater with CP, and seawater under free corrosion. The relationship between in-air and in-
                 seawater with cathodic protection (CP) varies between codes.  Using NORSOK (NTS, 1998),
                the fatigue life at high stress ranges (when N is less than106 cycles) in seawater with CP is
                 considered to be 40% of that in-air. However, there is no difference between the S-N curves at
                 lower stress ranges (when N is in excess of lo7 cycles).
                 In general, the fatigue life in seawater under free corrosion is 33% of the life in air at high
                 stress ranges  (when  N  is  less  than  lo7 cycles). There  is no  change  in  slope  for  the  free
                 corrosion S-N curve and hence the fatigue lives are around lo%, of the equivalent lives for in-
                 air S-N curve when N is more than lo7 cycles.
                 17.1.4  Effect of Mean Stress
                Compressive  mean  stress has  a  beneficial effect  on  fatigue capacity.  Normally  it  is  not
                required  to  account  for  the  effect  of  mean  stress. However, in  some  special cases,  it  is
                necessary to take into account the mean stress effect to modify the selected S-N curves, e.g.
                for the  fatigue assessment of TLP tethers and mooring lines whose non-linear response is
   350   351   352   353   354   355   356   357   358   359   360