MCNP Guide
MCNP Fundamentals
Core concepts before writing your first input
How MCNP simulates particles
MCNP tracks individual particle histories from source to termination using random sampling at each interaction — scatter, absorb, fission, or escape. Running millions of histories builds a statistically meaningful result with quantified uncertainty. This approach handles complex geometry and continuous-energy physics without the mesh or group-structure approximations that deterministic codes require.
Units and coordinate system
Cylindrical and spherical surface types (CZ, SO, etc.) are available, but all parameters are still specified in Cartesian coordinates — you cannot input (r, θ, z) directly.
Problem types
Fixed source
A known source with defined energy, position, and direction drives the calculation. Used for shielding design, dose calculations, and detector response. The source is specified with the SDEF card and histories are controlled by NPS.
Criticality (k-eigenvalue)
MCNP iterates neutron generations to converge the fission source and calculate k-effective. Controlled by the KCODE card (particles per cycle, inactive cycles, total cycles) and initial source points via KSRC.
Particles and nuclear data
MCNP transports neutrons, photons, and electrons (and combinations via the MODE card). Interaction probabilities come from evaluated nuclear data libraries (ENDF) processed into the continuous-energy ACE format. The ZAID extension identifies the library version and temperature — for example, .80c is ENDF/B-VIII.0 at room temperature. Thermal neutron scattering in bound materials requires additional S(α,β) data specified with MT cards.
For details on library extensions and cross-section selection, see Nuclear Data & Cross Sections.