DPM Physical Models Panel in ANSYS Fluent 2024 R1
#+CATEGORY: cfd
#+TAGS: DPM Fluent physical models forces thermophoretic
#+MATURITY: growing
DPM Physical Models :: Complete Panel Walkthrough for ANSYS Fluent 2024 R1
This page walks through every toggle and setting in the Fluent DPM Physical Models tab. Each option is explained with its governing physics and when it matters for PM2.5 aerosols. Target application: shisha simulation.
Accessing the Physical Models Tab
Models > Discrete Phase > Edit... > Physical Models tab
This tab controls:
- Which forces act on the particle (beyond drag and gravity)
- Whether heat/mass transfer is computed
- Whether particle-phase radiation is included
- Turbulence dispersion settings
Forces Section
Gravity (Body Force)
- Toggle:
Body Forcescheckbox - Default:
On - Direction: defined in
Operating ConditionsorCell Zone Conditions
Gravity always acts. For PM2.5 the gravitational settling velocity is negligible compared to turbulent diffusion (Stokes number << 1), but include it in the setup for completeness -- the cost is zero.
Saffman Lift Force
- Toggle:
Saffman lift forcecheckbox - Computes the lift force on a sphere in a shear flow:
: F_lift = 6.46 * mu * (rho_p / rho_f)^0.5 * d_p^2 * (omega x (u_p - u_f)) / (nu * Re_omega)^0.5
- Relevant for: large particles (> 10 \mu m) in strong shear layers
- For PM2.5: negligible. Saffman force scales with d_p^2, which is 10^-12 for 1 \mu m particles. You can leave it off or on -- either way the contribution is subdominant.
Virtual Mass Force
- Toggle:
Virtual masscheckbox - Accounts for the added mass effect: accelerating a particle also accelerates the surrounding fluid
: F_vm = -C_vm * rho_f * V_p * (du_p/dt - du_f/dt)
where C_vm = 0.5 for a sphere.
- Relevant for: bubbles in a liquid (rho_p < rho_f) or strongly accelerating flows
- For PM2.5 (solid aerosol in air): negligible. rho_p / rho_f ~ 1000, so the virtual-mass term is ~10^-3 of the drag.
Pressure Gradient Force
- Toggle:
Pressure gradient forcecheckbox - Force due to the pressure gradient in the flow:
: F_pg = -V_p * grad(p)
- Relevant for: bubbles, low-density particles in strong pressure gradients
- For PM2.5 in shisha: negligible. The pressure gradient across the chamber is small, and particle density is high relative to air.
Thermophoretic Force
- Toggle:
Thermophoretic forcecheckbox - Drives particles from hot to cold regions
: F_th = -D_T * grad(T) / T
where D_T is the thermophoretic coefficient (depends on particle size, thermal conductivities of gas and particle).
- Relevant for: combustion, hot exhaust, smoke cooling
- For shisha PM2.5: potentially significant. Smoke enters at ~350 K and cools toward ambient (~300 K). The temperature gradient near walls can drive sub-micron particles toward the glass. Enable this if you solve the energy equation and expect temperature gradients > ~10 K/cm.
How to compute thermophoretic force for specific particle sizes: Thermophoretic velocity scales approximately as: : v_th ~ (k_g / rho_p / d_p) * (grad(T) / T)
For d = 1 \mu m, grad(T) = 100 K/m, k_g = 0.026 W/mK, rho_p = 1200 kg/m^3: : v_th ~ 0.026 / (1200 * 1e-6) * (100 / 300) ~ 7 mm/s
Compare to typical flow velocity ~0.1 m/s: thermophoretic drift is ~7% of the flow. Marginal but not zero. Include it.
Heat and Mass Transfer Section
Enable heat transfer
- Toggle:
Heat transfercheckbox - When enabled, Fluent tracks the particle temperature and computes convective heat transfer using the Ranz-Marshall correlation
For inert PM2.5: enable this if you want any temperature change on the particle. The heating time constant is ~4 ms for 1 \mu m particles, so they reach thermal equilibrium with the local gas very quickly.
Evaporation (Law 2) / Boiling (Law 3) / Wet combustion
- These laws are only active for
DropletorMulticomponentparticle types - For
Inertparticles (shisha smoke): /all evaporation/boiling laws are inactive/ regardless of toggle state
Radiation
- Toggle:
Radiation Interactioncheckbox - Computes radiative heat exchange between particles and surrounding surfaces/medium
- For PM2.5: negligible at shisha temperatures. Smoke particles are too small to emit/absorb significant radiation compared to convective exchange.
Coupled heat transfer
- Toggle:
Coupled heat transfercheckbox (only with two-way coupling) - Enables the heat source term from particles back to the continuous phase
- For one-way coupling (shisha PM2.5): not applicable
Turbulence Dispersion Section
Stochastic Tracking (DRW)
- Toggle:
Stochastic trackingradio button - Activates the Discrete Random Walk model
- Parameters:
- Number of Tries: how many stochastic trajectories per parcel (default = 1)
- Eddy lifetime model:
default(eddy lifetime based on k-epsilon) or =TKE-based=/dissipation-based
This is mandatory for PM2.5. Without stochastic tracking, all particles follow the mean flow and there is zero turbulent dispersion. With DRW, the turbulent velocity fluctuations are sampled, and the particle cloud spreads realistically.
See also: DPM Turbulence Interaction for the full DRW algorithm.
Cloud Model
- Toggle:
Cloud modelradio button (alternative to stochastic tracking) - Represents the particle cloud as a Gaussian distribution (no individual trajectories)
- Used for: spray combustion where you need ensemble statistics without tracking every parcel
- For PM2.5: not recommended. The cloud model smears out the particle distribution; stochastic tracking gives pointwise trajectory information you need for concentration fields.
Additional Forces (User-Defined)
DEFINE\_DPM\_BODY\_FORCE
- Add a user-defined body force to the particle equation of motion
- This is the way to add Brownian motion for PM2.5
Shisha PM2.5 configuration:
- Enable
Thermophoretic force - Enable
Stochastic trackingwith Number of tries = 10 - Write a UDF for Brownian force using
DEFINE_DPM_BODY_FORCE - Leave
Saffman,Virtual mass,Pressure gradientunchecked (they are negligible but cost almost nothing; you can leave them on for completeness)
Full Physical-Models Checklist for Shisha PM2.5
| Setting | Value | Reasoning |
| --------- | ------- | ----------- |
| Gravity | On | Free; completeness |
| Saffman lift | Off | Negligible for d < 2.5 \mu m |
| Virtual mass | Off | Negligible (rho_p >> rho_f) |
| Pressure gradient force | Off | Negligible |
| Thermophoretic force | On | Temperature gradient ~ 10--50 K/cm near walls |
| Brownian force (UDF) | On via DEFINE\_DPM\_BODY\_FORCE | Sub-micron particles undergo Brownian diffusion |
| Heat transfer | On | Track particle temperature for completeness |
| Evaporation | Off | Inert particle type; laws are inactive |
| Radiation | Off | Negligible for PM2.5 at shisha temperature |
| Turbulence dispersion | Stochastic tracking | Mandatory for dispersed PM2.5 |
| Number of tries | 10 | Statistical convergence of turbulent spread |
--- :PROPERTIES: :seeAlso: cfd_dpm-setup-activation, cfd_dpm-turbulence-interaction, cfd_dpm-heat-mass-transfer, cfd_dpm-brownian, cfd_dpm-shisha-pm25 :END:
- DPM Brownian Motion -- Langevin Equation, Diffusion Coefficient, UDF for PM2.5CFD - DPM
- DPM Coupling -- One-Way vs Two-Way, Source Terms, and Decision LogicCFD - DPM
- DPM Drag Models -- All Drag Laws, Validity Ranges, and UDF ImplementationCFD - DPM
- DPM Erosion and Accretion Maps in ANSYS Fluent 2024 R1CFD - DPM
- DPM Heat and Mass Transfer -- Convective Heating, Evaporation, Boiling, and RadiationCFD - DPM
- DPM Setup and Activation in ANSYS Fluent 2024 R1CFD - DPM
- Shisha PM2.5 DPM Case Study -- Setup, Configuration, and ValidationCFD - DPM
- DPM Theory -- Governing Equations, Parcels, and Particle Force BalanceCFD - DPM