The cable-stayed portion of the Runyang bridge has a main span of 400 m supported between towers by 52 cables on each side. The cable length and location differ contributing to different tension that would vary depending on the traffic conditions. High stresses prevail in cables near the free ends of the cantilever while the low stresses occur near quarter way of the main span and at the middle of the bridge. Such a variation has the tendency of tightening and loosening the cables in a complicated manner. The ways with which this variation would affect the damage of the cables by fatigue crack growth becomes increasingly more important in time, particularly for developing the methodology of inspection and maintenance of the bridge cables the failure of which by fatigue is likely to be location specific. One of the objectives of this work is to develop a model that can systematically determine the fatigue crack growth in pre-tensioned bridge cables that are made of stranded steel wires. The idealized cable containing 100% solidly packed wires correspond to β = 1.0 in the model. This parameter is shown to have a significant effect on the fatigue strength of the cables when using a dual scale crack growth rate relation that accounts for both macro- and micro-effects in geometry and material property. The combined influence of mean stress and stress amplitude on crack growth is shown to depend on tightening and loosening of the cables and steel wires. Variation of the tension in the cable and/or wire can be assessed by a parameter α such that α = 1 (normal tension) can serve as the reference. More and less tension correspond, respectively, to α ⩾ 1 and α ⩽ 1 . In this way, the effect of traffic on fatigue can be delineated from that without the traffic. The results are presented graphically and discussed systematically for the cables subjected to high and low stresses. This provides an overall assessment of the fatigue crack growth behavior in the Runyang cable-portion of the bridge for a total of two million cycles. In general, it can be said that Varying max. stress in cable with traffic : cable fatigue life can be enhanced and impeded, respectively, by loosening and tightening of tension in cable. Varying max. stress in cable without traffic : cable fatigue life can be enhanced and impeded, respectively, by tightening and loosening of tension in cable. Since geometric symmetry of the 52 cables cannot be satisfied exactly about the middle of the bridge, the high stresses in cable #3 is only approximately equal to that in cable #50. The same applies to the low stresses in cable #14 when compared to that in cable #28. Because of the significant difference in the fatigue life of cable #50 and #28, it is worthwhile to use a higher modulus material for cables #3 and #50 in contrast to that for cables #14 and #28. This conclusion is based on using energy density function of the dual scale micro-/macro-crack growth model. Since the results depend on the choice of the fatigue crack growth criterion, validation of engineering application should rely on minimizing contradictions and inconsistencies. Moreover, multiple criteria may be considered for the future design of bridges similar to the design of airborne vehicles that incorporate the concept of “fail safe”.