DPM Boundary Conditions in ANSYS Fluent 2024 R1

#+CATEGORY: cfd

#+TAGS: DPM Fluent boundary conditions trap escape reflect

#+MATURITY: growing

DPM Boundary Conditions :: Complete Reference for ANSYS Fluent 2024 R1

This page covers every particle boundary condition Fluent supports, when to use each, and how they interact with the continuous-phase BCs. Target application: shisha PM2.5 simulation.

DPM BC Decision Matrix

How to set DPM BCs: Cell Zone Conditions > Boundary panel group > <zone-name> > DPM tab

DPM BCs by Continuous-Phase Boundary Type

Important: Interior zones cannot have DPM BCs assigned. Particles simply pass through interior faces.

Trap Boundary Condition (Deposition)

The particle terminates on the wall. Fluent records:

• Impact position (x, y, z)
• Impact velocity (u, v, w)
• Impact diameter and mass
• Impact time

Use trap when you want to quantify:

• Particle deposition on surfaces (soot, dust, PM2.5 on walls)
• Erosion rate (accumulated in erosion/accretion variable)
• Deposition patterns for filter design

For shisha PM2.5: Set wall zones to trap on interior chamber walls to quantify where smoke deposits accumulate. The trap BC records impact data that feeds into erosion/accretion reports.

Escape Boundary Condition

The particle exits the computational domain. It is counted in the DPM Summary as escaped. Use escape for:

• All outlet boundaries (pressure-outlet, outflow)
• All inlet boundaries (particles ride the inflow in)
• Far-field boundaries (open atmosphere)
• Symmetry axes where particles should pass through

Escape records:

• Exit position
• Exit velocity
• Exit diameter and temperature

Reflect Boundary Condition (Restitution)

The particle bounces off the wall with a coefficient of restitution. You specify two values:

• Normal restitution coefficient: ratio of post- to pre-impact normal velocity (0 = full trap, 1 = elastic normal)
• Tangential restitution coefficient: ratio of post- to pre-impact tangential velocity

Parameters:

• Normal restitution: typically 0.2--0.5 for soot/smoke on glass
• Tangential restitution: typically 0.5--0.9

Use reflect when particles genuinely bounce: sand on metal, granular flow on walls, water droplets on hydrophobic surfaces.

For shisha PM2.5: If smoke particles deposit on walls (stick), use trap. If they bounce (e.g., polished glass bowl), use reflect with low normal restitution (~0.1--0.2).

Bounce (Elastic)

Elastic impact: normal and tangential restitution both equal 1.0. The particle bounces as if it hit a perfectly elastic surface. Used for:

• Idealized simulations (no energy loss)
• Debugging trajectories (if bounce is wrong, check the wall geometry)

Not appropriate for real PM2.5 deposition.

Wall-Jet

After impact, the particle continues along the wall surface, driven by the tangential component of its impact velocity plus the local fluid shear. Use wall-jet for:

• Erosion simulation where the particle scrapes along the surface
• Sand-blast modeling
• Liquid-film formation from impacting spray

Semi-Reflect (Droplet-Specific)

Available only for Droplet particle type. The particle either bounces or spreads (forms a wall film) depending on the impact Weber number:

: We = rho * V^2 * d / sigma_s

• We < We\_critical: droplet bounces
• We > We\_critical: droplet spreads into a wall film

The critical Weber number is user-adjustable (default ~100 for water on dry surfaces).

Not applicable to inert PM2.5 (shisha smoke).

Interior and Periodic Boundaries

Interior zones: Particles always pass through interior faces. You cannot assign a DPM BC to an interior zone. This is correct behavior -- interior faces are internal to the mesh, not physical boundaries.

Periodic boundaries: Particles exit one periodic face and re-enter the matched face. This is essential for simulations with periodic geometry (e.g., a repeating filter element).

Mapped boundaries (non-conformal interfaces): Particles pass through mapped interfaces. Verify the grid-connectivity is correct, or particles may be lost.

BCs and the Continuous Phase

The DPM BC does not have to match the continuous-phase BC. Common mismatched setups:

• Wall (no-slip for fluid) + trap for particles (wall stops fluid, captures particles)
• Pressure-outlet (fluid leaves) + escape (particles leave too)
• Wall + reflect (fluid has no-slip, particles bounce)

The DPM BC controls only the particle behavior. The fluid BC controls the continuous phase independently.

Shisha PM2.5 Boundary Condition Strategy

DPM Summary Report and BC Accounting

After DPM iterations converge, Report > Discrete Phase > Summary prints a table:

Mass balance check: injected should equal the sum of escaped + trapped + evaporated (for inert particles, evaporated = 0). If incomplete > 0, particles are stuck (too many steps or zero-velocity cells). Increase max steps or check mesh quality.

--- :PROPERTIES: :seeAlso: cfd_dpm-setup-activation, cfd_dpm-injection-types, cfd_dpm-physical-models, cfd_dpm-numerics, cfd_dpm-shisha-pm25 :END: