Density dependence of the fan curve (HVAC model)

[Prometeo-ianic]

Hi all,
I'm modeling a fan using the HVAC system (the FAN namelist with a pressure–volume flow curve defined via RAMP). The performance curve I provide is of course referenced to a specific air density (typically standard conditions, ~1.2 kg/m³).
In a fire scenario, however, the density of the gas passing through the fan can change significantly due to the temperature rise. Per the well-known fan laws, at constant rotational speed the pressure rise scales roughly linearly with density (Δp ∝ ρ), while the volumetric flow rate stays essentially unchanged.
My question is: does FDS account for this automatically? That is, does the HVAC solver correct the fan curve as a function of the local density (or temperature) at the fan node, or does it apply the curve exactly as defined, regardless of the conditions of the gas flowing through it?
In other words, if the fan is moving hot, low-density smoke, does the code reduce the effective pressure rise, or do I need to manually correct the RAMP for the expected conditions?
Thanks in advance for any clarification, including any reference to the relevant section of the Technical Reference Guide.

[drjfloyd]

While density is accounted for in the 1D momentum equation used to set the pressures on either end of a duct, FDS is not shifting the fan curve. Good comment and something I'll look into.

[drjfloyd]

Added the density fan affinity law and a verification case in this pull request: #16370

New input (CURVE_TEMP) for HVAC FAN added to set the assumed temperature for the fan curve density (default is 20 C).

[Prometeo-ianic]

Thank you very much for taking the time to investigate this. I really appreciate . I'll test it as soon as possible . Thanks again!