A recently published two-dimensional, two-group hexagonal geometry IAEA-Hex kinetics benchmark was solved using the nodal neutronics programs Ants and DYN3D. Solutions to both the ramp and step transients were obtained with both programs. With Ants, the benchmark was solved using either the hexagonal geometry model or the triangular geometry model with either 6 or 24 triangles per hexagon. For DYN3D, three different hexagonal nodal models were used to solve the benchmark. For both transients and all models, multiple solutions with different fixed time-step lengths were calculated.

The results were compared against the benchmark reference results. With Ants, the triangular geometry model predicted the powers more accurately than the hexagonal model. Refining the radial mesh with the triangular geometry model did not affect the results of either transient. Generally, Ants agreed very well with the reference solution when a short enough time step was used, especially with the triangular geometry model. DYN3D yielded results that were sufficiently close to the reference solution, demonstrating its capability to accurately model transient phenomena in hexagonal reactor cores. The sensitivity of the solution accuracies and run times to the time step used in the transient calculations was also investigated.