PSI - Issue 83

Iman Onsa et al. / Procedia Structural Integrity 83 (2026) 273–279

276

2.4. Boundary conditions The casing was fully constrained, while the TJ was rotating at a constant speed of 115RPM. Side load was applied as a normal force between the TJ and casing to represent contact loading in directional drilling. Simulations were conducted at 1000N, 1200N and 1400N, the same as in the experiment, as shown in Error! Reference source not found. . 2.5. Contact and friction modelling (OBM vs WBM) Automatic surface-to-surface contact was defined between the TJ outer surface and the inner surface of the casing using explicit contact formulation. To represent drilling fluid lubricity, contact COF was assigned for OBM of 0.18 and 0.23 for WBM cases. All model parameters were kept identical between OBM and WBM runs except for friction, enabling a direct comparison of lubrication effects. 2.6. Wear modelling (Archard wear law) Wear was computed using Archard’s wear law, where local wear is proportional to contact pressure and sliding distance. The model used in the experiment determined wear coefficients to calibrate wear predictions under standard test conditions, then applied the calibrated setup across the OBM/WBM and load cases. The expected output of the law includes the wear depth distribution, total wear volume, and specific wear rate [7]. ̇ = ′ ∥ t ∥ or Δℎ= ′ Δ = The simulation was done for 1s, then the wear results were extrapolated to the full experimental time of 190 minutes. Model predictions were compared against available experimental wear data [8] under the same conditions (115RPM, 1000N) to confirm that wear location and magnitude were reasonable. After the validation, the same modelling settings were used to evaluate OBM vs WBM at different load ranges to ensure consistency in the comparison. 3. Results 3.1. Validation results The calibrated FE wear model reproduced the expected wear localization at the casing-TJ contact region and provided wear depth and volume values consistent with the experimental wear data. The results are shown in the figure, which supports using the model to compare WBM at 115 RPM and 1000N. Figure 2 shows the validation 2.7. Validation and scaling