Abstract: The small lead-cooled fast reactor (LFR), as a core reactor type in the fourth-generation advanced nuclear energy system, boasts advantages such as compact structure and outstanding inherent safety, with broad application prospects in the energy sector. However, its small core size and densely packed fuel assemblies lead to strong leakage effects and pronounced spatial effects during transient processes. Additionally, the fast reactor’s neutron spectrum is relatively hard, exhibiting pronounced anisotropic neutron scattering and notable neutron breeding effects. Traditional diffusion calculation methods based on the P0 isotropic approximation are prone to introducing systematic biases, failing to meet the requirements for core design and safety analysis. Based on the two-step framework, the steady-state module of the independently developed MCDESOF1.0 program by the China Institute of Atomic Energy was extended to create the MCDESOF2.0 program, which featured 3D spatiotemporal dynamics solving and external source handling capabilities. Addressing the characteristics of the strong leakage system in small lead-cooled fast reactors, the Monte Carlo program OpenMC was employed to generate multigroup cross-section constants. By incorporating transport corrections, (n, xn) reaction corrections, and an improved super homogenization (SPH) correction method to optimize cross-section equivalence accuracy, the program’s stability and computational efficiency were further enhanced. To verify the reliability and applicability of the developed MCDESOF2.0 program, this study utilized historical reaction disturbance experimental data on the lead core of the YBS critical device at the China Institute of Atomic Energy, including critical state control rod and safety rod drop experimental data, as well as subcritical state safety rod and two sets of regulating rod lifting experimental data. The 2# regulating rod experiment was conducted with the 1# regulating rod inserted into the core, forming different subcritical conditions and providing experimental benchmark data for subsequent numerical simulation verification. The MCDESOF2.0 program was employed to conduct full-process numerical simulations for all experimental conditions, systematically comparing the simulation results with experimental values. The findings indicate that after corrections, the effective reactivity coefficients of each rod position are controlled within 20 pcm, with an RMS deviation of approximately 0.30% in the full-core power distribution. The calculated reactivity trends from critical rod drop experiments are generally consistent with the experimental results, showing a maximum detector current signal RMSE of 9.67% and a maximum reactivity NRMSE of 13.7%. Under subcritical conditions, the calculation results for both control rod withdrawal and insertion processes align well with experimental measurements, the relative deviation between the calculation results of the main parameters and the experimental values is basically within ±5%, and the relative deviation at some points is about ±10%. In summary, the established computational method effectively describes the core dynamic response characteristics in transient experiments of small LFR, providing a reference for transient behavior analysis of small fast reactors and the validation of related numerical programs.