Catalytic Converter: Flow Optimisation Study

What Is This Study About?

A catalytic converter works by passing exhaust gas through porous catalyst blocks where chemical reactions neutralise pollutants. The efficiency of those reactions depends on how evenly the flow is distributed across the catalyst face, a concentrated jet hitting one region and starving another means uneven catalyst loading, faster localised degradation, and reduced conversion efficiency.

This study uses SolidWorks Flow Simulation to evaluate six design configurations, starting from a bare baseline and progressively refining a guided vane geometry, to find the configuration that simultaneously produces the most uniform catalyst inlet flow and the lowest total pressure drop.

Two porous media blocks represent the catalyst sections. The study uses cold-flow air at a fixed 10 m/s inlet velocity so that all cases are directly comparable.

Physical Setup & Geometry

The converter geometry contains:

Inlet pipe, 10 m/s velocity inlet, air cold-flow
Two isotropic porous catalyst blocks, pressure-drop and flow-distribution targets
Outlet, static pressure at 101,325 Pa

The geometry was created in SolidWorks and the flow study set up as an internal analysis. Catalyst blocks are modelled as porous media rather than resolving individual honeycomb channels, this keeps the model computationally light while preserving the pressure-drop and flow-distribution trends needed for design comparison. All cases are run at basic mesh settings for quick evaluation.

Baseline converter geometry, front view — Baseline geometry, front view. Two porous catalyst blocks downstream of the inlet pipe with no additional flow-control feature.

Baseline converter geometry, isometric view — Baseline geometry, isometric view.

Simulation Setup

Parameter	Value
Analysis type	Internal flow
Fluid	Air, cold-flow
Inlet condition	Velocity inlet, 10 m/s for all cases
Outlet condition	Static pressure, 101,325 Pa
Catalyst approximation	Two isotropic porous media blocks
Porous media reference area	20,758 mm² (0.020758 m²)
Porous media reference length	100 mm (0.1 m)
Primary metric	Velocity uniformity ratio at Porous Block 1 front face
Secondary metric	Total converter pressure drop

Goal Equations

Both comparison metrics are equation goals extracted using the Averaged Value column:

Velocity uniformity ratio = Max velocity at Porous 1 front face ÷ Average velocity at Porous 1 front face
Total converter pressure drop = Inlet average static pressure − Outlet static pressure

The uniformity ratio is dimensionless, velocity units cancel in the division. A ratio of 1.0 is perfect uniformity; higher values indicate greater maldistribution.

Design Case Overview

Case	Configuration	Uniformity Ratio	vs Baseline	Total DP (Pa)	vs Baseline	Decision
1	Baseline, no baffle	2.832	,	114.065	,	Reference
2	Circular baffle	3.587	−26.7%	112.079	−1.7%	Rejected
3	Guided vane, 4 mm thick, 30 mm wide, 80 mm	2.557	+9.7%	106.901	+6.3%	Improved
4	Guided vane, 4 mm thick, 20 mm wide, 90 mm	2.602	+8.1%	110.016	+3.5%	Improved
5	Guided vane, 8 mm thick, 30 mm wide, 80 mm	1.990	+29.7%	96.588	+15.3%	Improved
6	Guided vane, 8 mm thick, 20 mm wide, 90 mm	1.540	+45.6%	93.120	+18.4%	Best

Uniformity ratio: lower = better. Pressure drop: lower = better.

Case 1, Baseline: No Baffle

The unmodified converter geometry with two porous catalyst blocks and no additional flow-control feature.

Uniformity ratio: 2.832 · Total DP: 114.065 Pa · P1 avg/max velocity: 1.203 / 3.406 m/s

The inlet jet reaches the first catalyst face with a concentrated high-velocity core. The ratio of 2.832 confirms that peak velocity is nearly three times the face-averaged value, strong maldistribution.

Case 1, velocity cut plot — Velocity cut plot, Case 1. The inlet jet penetrates directly toward the first catalyst block with no redirection.

Case 1, static pressure cut plot — Static pressure cut plot, Case 1. Pressure gradient concentrated at the catalyst inlet face.

Case 1, front face velocity distribution at Porous Block 1 — Velocity distribution at Porous Block 1 front face, Case 1. The high-velocity core at the centre is clearly visible against low-velocity peripheral regions.

Case 1, flow trajectories coloured by velocity — Flow trajectories coloured by velocity, Case 1. Streamlines concentrate through the centre of the catalyst, confirming the maldistribution.

Case 2, Circular Baffle

A circular baffle placed upstream of the first porous block to block the inlet jet directly.

Uniformity ratio: 3.587 · Total DP: 112.079 Pa · P1 avg/max velocity: 1.120 / 4.016 m/s

Case 2, circular baffle geometry, front view — Case 2 geometry, circular baffle positioned upstream of Porous Block 1.

Case 2, circular baffle geometry, isometric view — Isometric view, circular baffle concept.

Case 2, velocity cut plot — Velocity cut plot, Case 2. The baffle forces flow around its edges, creating accelerated peripheral jets that worsen the distribution.

Case 2, front face velocity distribution — Front face velocity distribution, Case 2. Peak velocity increased to 4.016 m/s, worse than baseline. Concept rejected.

Case 2, flow trajectories coloured by velocity — Flow trajectories, Case 2. Blocking the jet redirects flow to the periphery, creating edge-concentrated high-velocity paths through the catalyst.

Rejected. Simply blocking the inlet jet accelerates flow around the baffle edges and increases the uniformity ratio to 3.587, worse than the unmodified baseline. This confirms that jet-blocking without redirection is counterproductive.

Case 3, Guided Vane: 4 mm Thick, 30 mm Wide, 80 mm Location

First guided vane concept. The vane redirects the inlet jet gradually rather than blocking it.

Uniformity ratio: 2.557 · Total DP: 106.901 Pa · P1 avg/max velocity: 0.752 / 1.924 m/s

Case 3, guided vane geometry, front view — Case 3 geometry, 4 mm thick, 30 mm wide guided vane at the 80 mm location.

Case 3, guided vane geometry, isometric view — Isometric view, first guided vane concept.

Case 3, velocity cut plot — Velocity cut plot, Case 3. The vane redirects the jet, spreading momentum more broadly before the catalyst face.

Case 3, front face velocity distribution — Front face velocity distribution, Case 3. Max velocity drops to 1.924 m/s vs 3.406 in the baseline. Improved.

Case 3, flow trajectories — Flow trajectories, Case 3. The guided vane spreads flow across a wider area of the catalyst compared to the baseline jet.

Case 4, Guided Vane: 4 mm Thick, 20 mm Wide, 90 mm Location

Narrower 4 mm vane placed closer to the catalyst inlet.

Uniformity ratio: 2.602 · Total DP: 110.016 Pa · P1 avg/max velocity: 0.935 / 2.434 m/s

Case 4, guided vane geometry, front view — Case 4 geometry, 4 mm thick, 20 mm wide guided vane at the 90 mm location.

Case 4, guided vane geometry, isometric view — Isometric view, narrower 4 mm vane positioned closer to the inlet.

Case 4, velocity cut plot — Velocity cut plot, Case 4. Narrower vane provides less jet redirection than Case 3.

Case 4, front face velocity distribution — Front face velocity distribution, Case 4. Ratio of 2.602, slightly worse than Case 3. Reducing width at this location was not beneficial.

Case 4, flow trajectories — Flow trajectories, Case 4. Narrower vane captures less of the inlet jet, leaving more maldistribution than the wider Case 3 vane.

Case 5, Guided Vane: 8 mm Thick, 30 mm Wide, 80 mm Location

Increasing vane thickness to 8 mm while keeping the 30 mm width and 80 mm location from Case 3.

Uniformity ratio: 1.990 · Total DP: 96.588 Pa · P1 avg/max velocity: 0.434 / 0.863 m/s

Case 5, guided vane geometry, front view — Case 5 geometry, 8 mm thick, 30 mm wide guided vane at 80 mm. The thicker cross-section provides stronger jet redirection.

Case 5, guided vane geometry, isometric view — Isometric view, 8 mm thick guided vane.

Case 5, velocity cut plot — Velocity cut plot, Case 5. The thicker vane produces notably more uniform flow across the catalyst inlet face.

Case 5, front face velocity distribution — Front face velocity distribution, Case 5. Max velocity reduced to 0.863 m/s. A large step improvement over all previous cases.

Case 5, flow trajectories — Flow trajectories, Case 5. The thicker vane spreads the inlet jet across the full catalyst face much more effectively.

Case 6, Guided Vane: 8 mm Thick, 20 mm Wide, 90 mm Location (Final Design)

Reduced width and shifted the vane closer to the inlet. This is the best performing configuration.

Uniformity ratio: 1.540 · Total DP: 93.120 Pa · P1 avg/max velocity: 0.412 / 0.635 m/s

Case 6, final guided vane geometry, front view — Case 6 geometry, 8 mm thick, 20 mm wide guided vane at the 90 mm location. Final optimised configuration.

Case 6, final guided vane geometry, isometric view — Isometric view, final guided vane. The narrower width at closer proximity to the inlet provides the best jet-redirection balance.

Case 6, velocity cut plot showing highly uniform flow — Velocity cut plot, Case 6. The most uniform flow distribution of all six cases. The inlet jet is fully redirected and distributed before reaching the catalyst.

Case 6, static pressure cut plot — Static pressure cut plot, Case 6. Total pressure drop reduced to 93.120 Pa, the lowest of all cases, including the baseline.

Case 6, front face velocity distribution showing near-uniform loading — Front face velocity distribution, Case 6. Max velocity 0.635 m/s vs 3.406 in the baseline. Near-uniform loading across the full catalyst face.

Case 6, flow trajectories coloured by velocity — Flow trajectories, Case 6. Streamlines distribute evenly across the catalyst, confirming the maldistribution has been resolved.

Results & Engineering Conclusions

Key findings

Blocking a jet is not the same as redirecting it. The circular baffle (Case 2) worsened flow distribution to a ratio of 3.587, higher than the 2.832 baseline, because it forced flow around its edges, creating accelerated peripheral jets. Guided vanes that redirect rather than obstruct were consistently superior.

Vane thickness is the dominant design variable. Doubling thickness from 4 mm to 8 mm (Cases 3→5) reduced the uniformity ratio by 22% and cut total pressure drop by 9.6 Pa. Width and axial position produced smaller but still measurable improvements.

The final design improves both objectives simultaneously. The 8 mm guided vane (Case 6) achieved a 45.6% improvement in uniformity ratio and an 18.4% reduction in total pressure drop compared to baseline, demonstrating that flow redirection does not require a pressure-drop penalty when the vane geometry is well-tuned.

Porous media approximation is valid for design comparison. Mass flow balance was confirmed in every case (inlet and outlet magnitudes match). The porous media model is appropriate for relative trend analysis; absolute quantitative results would require directional resistance coefficients and mesh sensitivity validation before use in final design decisions.

Final Design Comparison

Metric	Baseline (Case 1)	Final Design (Case 6)	Improvement
Velocity uniformity ratio	2.832	1.540	45.6% lower
Total converter pressure drop	114.065 Pa	93.120 Pa	18.4% lower
Porous Block 1 max velocity	3.406 m/s	0.635 m/s	81.4% lower
Porous Block 1 avg velocity	1.203 m/s	0.412 m/s	65.8% lower

Numerical Results, All Cases

Case	Inlet MF (kg/s)	Outlet MF (kg/s)	P1 Avg V (m/s)	P1 Max V (m/s)	Uniformity	DP P1 (Pa)	DP P2 (Pa)	Total Porous DP (Pa)	Total DP (Pa)
1	0.006056	−0.006056	1.203	3.406	2.832	2.528	3.641	6.205	114.065
2	0.006056	−0.006056	1.120	4.016	3.587	2.504	3.647	6.187	112.079
3	0.006055	−0.006055	0.752	1.924	2.557	3.498	3.659	7.193	106.901
4	0.006055	−0.006055	0.935	2.434	2.602	3.114	3.655	6.805	110.016
5	0.006054	−0.006054	0.434	0.863	1.990	3.874	3.658	7.569	96.588
6	0.006054	−0.006054	0.412	0.635	1.540	3.882	3.658	7.576	93.120

Limitations & Next Steps

Isotropic porous media, real catalytic converter honeycombs are directionally porous, with low resistance along channels and high resistance radially. A directional resistance model would give more accurate absolute pressure-drop values.
Cold-flow study, thermal effects, exhaust gas composition, transient pulsation from the engine, and catalytic heat generation are not included. A hot-flow study would be needed for thermal sizing.
Mesh sensitivity, current results are most reliable for relative comparison. A mesh sensitivity study comparing global and locally refined settings should precede any use of absolute values for final design decisions.
Future directions, directional porous media, elevated flow rates, thermal coupling, and a broader geometric sweep of vane angle and position.