This industrial fan CFD tutorial covers the closed duct : cylindrical domain, dedicated presets, Phase 2 under close watch.

Start with tutorial no. 1

A confined axial machine — duct fan, bulb turbine, propeller pump, ducted thruster — cannot be calculated like a propeller in free air. The entire flow rate passes through the duct, the duct wall is part of the problem, and the blade tip clearance becomes a first-order quantity. This tutorial shows what that costs, what it delivers, and where the current chain reaches its limits — because this is the hardest configuration on the test bench, and it's better to know that before launching a two-hour run.

• Prerequisites : a ducted propeller in Heliciel — here, the shipped fan model ;
• Duration of the case shown here : 1 h 35 min of meshing at the lightest preset, 3 min 35 s for Phase 1 ;
• What you get out of it : a readable pressure field inside the duct, and the ability to recognize a duct mesh that won't hold.

• Tutorial outline:

1. Four configurations, not one
2. The machine under study, and its BEM reference
3. The cylindrical domain and its cost
4. The duct presets
5. The wall conditions applied
6. Reading the result — and putting it in perspective
7. Recognizing a duct mesh that won't hold
8. Phase 2 in a duct, and its warning

1: Four configurations, not one

The bench distinguishes four case families, and the choice is made at creation :

• wing or airfoil : box-shaped domain, steady-state solver ;
• free propeller : rotor with no duct, box domain, MRF Phase 1 then sliding-mesh Phase 2 ;
• ducted propeller in free stream : the duct is there, but the fluid can flow around it — Kort nozzle, drone thruster ;
• closed duct : all of the flow rate passes through the machine, the duct wall closes off the domain. This is the subject of this page.

When you create the case from a ducted propeller, Heliciel asks the question : free stream or duct ? Answer duct for a duct fan, a penstock turbine, a propeller pump. The answer changes the domain, the boundary conditions and the preset catalogue — it cannot be corrected afterwards.

2: The machine under study, and its BEM reference

The fan model shipped with Heliciel is used as the example. Like every shipped model, it opens with a design point at zero : run Analyze operating point before switching to CFD, otherwise the BEM reference in the case folder will read zero.

An efficiency of 0,206 is not an anomaly : for a fan, the "propulsive efficiency" in the propeller sense doesn't carry the meaning it has on an aircraft. What matters for a duct machine is the pressure/flow-rate coupling — and it's the pressure rise, 103 Pa here, that the CFD calculation needs to reproduce.

3: The cylindrical domain and its cost

In a closed duct, the domain is no longer a box : it is a cylinder meshed as an O-grid, five blocks, with an inlet, an outlet and a duct wall. The machine's geometry is inserted into it along with its duct. Direct consequence : the mesh is much heavier than for a free propeller, at a nominally equivalent preset.

For our fan, at the Duct Level 1 preset — the one called "Smoke: quick test" — :

For comparison, the lightest preset for a free propeller meshes in 50 seconds for 15 000 cells. The ratio is a hundred to one. Plan your time accordingly : in a closed duct, there is no "just to see" preset. The lightest one in the catalogue is already an evening-long run.

4: The duct presets

The catalogue changes with the configuration. In a closed duct, there are only four levels, and they are much more closely spaced than the eight levels for the free propeller :

Two useful remarks. Levels 3 and 4 share the same mesh geometry : what separates them is the care taken over snapping and layer-stack convergence, not the resolution. And the background grid doesn't change from one level to the next : in a duct, most of the cost comes from refinement around the machine, not from the domain.

5: The wall conditions applied

The results panel displays the boundary conditions actually applied, and this is a habit worth adopting in a closed duct — they are not the same as for a free propeller :

The frictionless wall is a deliberate choice of the chain : stacking boundary layers on the duct wall, on top of those on the blades, produces cells that cross over at the tip clearance in this context and makes the mesh fail. That's the reason for the next section. It comes at a price : head losses from wall friction are not modelled. For a fan, that means the calculation gives you the machine, not the ductwork network.

6: Reading the result — and putting it in perspective

The fan inside its duct, flow lines and surface pressures. The pressure rise can be read on either side of the rotor plane : downstream in red, upstream in green.

Surface pressures, 3D model hidden. As everywhere in the bench : uncheck the 3D model to see the field, otherwise the grey STL draws over it.

The forces, on the other hand, call for caution. For our case :

The bench provides a second, independent line of evidence here, separate from the wall forces : it compares the torque read on the surfaces with the torque derived from the angular momentum balance of the fluid crossing the rotor. The two should coincide. They give 52,2 N·m against 34,0 N·m, a ratio of 1,53 where 1 is expected — and the bench flags it as a warning. A gap like this cannot be fixed with a solver setting : it means the mesh isn't faithfully transporting momentum through the rotor zone.

Thrust doubled, torque cut by a third : these two gaps don't point the same way, and that's the sign this isn't a simple resolution shortfall. The field even contains a maximum velocity of 565 m/s for a machine whose blade tip turns at 64 m/s — a local extremum, produced by a handful of degenerate cells, and it's enough to throw off any automatic colour scale. This result is a result of the configuration, not a result of the machine : it should not be quoted as the fan's performance.

The results panel in a closed duct : it recalls the applied wall conditions before giving the forces. Read those first.

7: Recognizing a duct mesh that won't hold

The bench analyzes checkMesh and returns a reliability score. For our mesh, 52 %, with the following detail :

The two rows that matter are the skewness and the near-degenerate cells. A cell with a near-zero determinant has an effective volume close to zero : gradients blow up there, and that's where the 565 m/s comes from. The bench's diagnostic is unambiguous on this point — it flags a structural divergence in Phase 2.

These flaws concentrate in a known area : the blade tip clearance, where the mesh has to fit cells between the rotating blade and the duct wall, over a gap only a few millimetres thick. It's the hard point of the whole configuration, and it's the subject of ongoing work : if your duct case produces these signatures, it is not a mistake on your part.

Vertical cut of the axial velocities inside the duct : the flow is confined, it has nowhere to expand — that's the whole difference with a free propeller.

Cylindrical cut at 40 % of the radius : the unrolled surface slices through all seven blades and shows the duct along its full length.

8: Phase 2 in a duct, and its warning

Sliding-mesh Phase 2 is possible in a closed duct, but it requires a healthy mesh. With a skewness of 7 and cells of near-zero volume, the time step collapses within the first few revolutions and the run stalls : that's what the bench's diagnostic warns of, and it's right to say so before rather than after a two-hour run.

The recommended course of action, in order :

1. read the reliability score and the checkMesh detail before launching Phase 2 ;
2. if near-degenerate cells are flagged, don't launch : remesh at the next level up, which takes more care over snapping convergence ;
3. the "problem cells" overlay in the Show menu highlights very small-volume cells in red and writes a full report to the log — it's the tool that tells you where the mesh gave way ;
4. if the tip clearance is the cause, the Level 3 preset deliberately trims it to keep the mesh healthy : that's a trade-off, not a flaw.

For confined machines where Phase 2 succeeds, everything described in tutorial no. 5 applies unchanged. And for an energy-capturing machine in a duct — bulb turbine, Kaplan, penstock — the sign reading is the same as in tutorial no. 14 : positive torque, a machine that produces.

The collection of twenty CFD tutorials

« Previous: Wind turbine and tidal turbine  |  Next: Aircraft propeller, cruise and static thrust »

1. First CFD propeller simulation
2. Reading the OpenFOAM case folder
3. Choosing the mesh preset
4. The MRF zone and Phase 1
5. Phase 2 with sliding mesh
6. Comparing BEM and CFD
7. Reading the convergence of a run
8. Turbulence models
9. Boundary layer and y+
10. Reading a pressure map
11. Where the forces come from
12. Checking an airfoil polar
13. Marine propeller and cavitation
14. Wind turbine and tidal turbine
15. Fan in a closed duct (you are here)
16. Aircraft propeller, cruise and static thrust
17. The optimisation loop
18. Calibrating on a reference
19. From the hull to the propeller
20. The deliverable calculation file

This series accompanies the Heliciel design tutorials, which cover the BEM side: blade design, choosing the rotation speed, performance curves. CFD comes after them, to verify and to see.