Comparative Prediction of the Fatigue Life of Traction Arms on the New Holland TT75 Tractor in Chad: Miner, Chaboche-Lemaitre and Tikri Approaches

This study presents a comparative fatigue-life assessment of two traction-arm designs used on the New Holland TT75 tractor operating under the demanding agricultural conditions found in Chad: the current production arm (BA) and a geometrically optimized alternative (BP), both manufactured from S235 steel. Finite element simulations performed in ABAQUS/CAE 2017 were used to determine the maximum and equivalent stresses generated in each arm under applied pressures ranging from 90 to 125 MPa, the equivalent stresses being obtained through the Goodman mean-stress correction. Fatigue life was then predicted using three complementary damage models of increasing complexity: Miner's linear cumulative-damage law, the Tikri model, and the Chaboche-Lemaître continuous damage model, applied first to constant-amplitude loading and then to loading sequences that include periodic overload cycles (one overload every 50 cycles, with overload ratios R = 1.2, 1.4 and 1.6) representative of the irregular soil, stones and roots encountered during ploughing and traction work in Chadian fields. Reference life values were obtained by digitizing a published Wohler S-N curve for S235JR steel. Across every applied pressure and every overload scenario, the optimized arm (BP) developed systematically lower equivalent stresses and achieved a longer predicted fatigue life than the current arm (BA), regardless of the damage model used. Among the three models, the Chaboche-Lemaître formulation consistently gave the most accurate and most conservative life estimates, capturing the progressive, sequence-dependent nature of damage accumulation more realistically than Miner's rule, which tended toward optimistic, non-conservative predictions. These findings confirm that modest geometric optimization of the traction arm can meaningfully extend component durability under the harsh tropical operating conditions typical of Chadian agriculture, and they point to the Chaboche-Lemaître model as the preferred tool for life prediction in this application, pending confirmation through physical fatigue testing.