Skip to content

Regenerative-cooling model

Scope

The cooling module estimates steady one-dimensional wall heat transfer and coolant pressure loss along a supplied nozzle contour. It is intended for early channel sizing and sensitivity studies.

Channel count, width, height, land width, hot-wall thickness, cooled length, coolant mass flow and inlet state are explicit inputs. EnSim does not infer a channel design from thrust or impose an undocumented heat-flux target. The GUI uses a stated three-point converging-diverging contour for this standalone screening analysis; importing a detailed contour is not yet supported.

Gas-side convection

EnSim implements the standard Bartz correlation using chamber pressure, characteristic velocity, throat diameter, local area ratio, gas viscosity, specific heat, Prandtl number and the property-variation factor. Molecular weight and gamma come from the engine calculation when invoked through the GUI.

The correlation was derived for rocket thrust chambers and is sensitive to the choice of reference properties. NASA comparisons have documented substantial variation between Bartz-type estimates and measurements, particularly away from the throat; it is a correlation, not a universal heat-flux law.

Coolant side and pressure loss

The hydraulic diameter is computed from channel geometry. For turbulent flow, the coolant coefficient uses Gnielinski and smooth-channel Darcy friction uses a logarithmic correlation. Invalid Reynolds/Prandtl domains are rejected rather than silently switching fluids or correlations. Coolant bulk temperature is marched counter to the gas flow and the energy rise uses the actual channel mass flow.

The lower-level reduced thermal-profile API accepts contraction ratio and both conical half-angles explicitly. It is retained for equation-level studies; the desktop Cooling workspace uses the channel-resolved analysis described above.

Thermal resistance

At each station, gas convection, wall conduction and coolant convection are combined in series. Reported hot-wall and coolant-wall temperatures are the steady solution of that local network. Differences from the nominal material melting point and coolant critical temperature are diagnostic temperature differences, not certified safety margins or phase-stability predictions.

Not represented

Boiling, supercritical property tables, film cooling, rib roughness, manifold maldistribution, coking, conjugate axial conduction, radiation, transient thermal stress, creep and fatigue are outside the current model.

References