CFD Analysis of Dirty-Air Reduction with Diverse Rear Wing Topologies

Abstract

This study investigates how rear-wing topology influences wake quality and dirty-air generation in a tandem configuration using numerical simulations in STAR-CCM+. A series of wing geometries with varying endplate angles and size ratios were analysed to understand their impact on aerodynamic loading, flow coherence, and overall system efficiency.Results show that increasing suction on the leading wing improves its aerodynamic performance but simultaneously intensifies the wake, creating a stronger low-energy region that degrades the follower’s inflow quality. In contrast, moderate endplate angling up to 8 degrees and controlled downsizing were found to narrow the wake corridor, enhance total-pressure recovery, and reduce dirty-air severity. The findings highlight that no single geometry can optimise both wings simultaneously: designs favouring the leader tend to compromise the follower. Configurations that balance geometry and wake management, however, achieve cleaner inter-element flow and improved overall efficiency. These insights provide a foundation for future aerodynamic co-optimisation of multi-element systems where wake control and stability are critical.

Introduction

Aerodynamics plays a vital role in a Formula car’s performance, primarily governed by two forces — downforce and drag. Downforce increases tire grip and cornering ability, while drag resists motion and reduces top speed. Engineers aim to maximize downforce while minimizing drag, particularly through optimization of the rear wing, which contributes over twice as much downforce as the front wing and operates in relatively clean airflow.

In modern Formula 1, “dirty air” — the turbulent wake generated by a leading car — poses a major challenge, as it reduces the following car’s downforce and stability. Studies show that at 20 m separation, a trailing car loses about one-third of its downforce, increasing to nearly half at 10 m. The 2022 F1 regulations aimed to counter this by simplifying aerodynamics and producing cleaner wakes. However, the rear wing remains a dominant source of turbulence due to its pressure differentials and tip vortices.

The present study investigates how rear-wing geometry influences wake behaviour and the aerodynamic performance of trailing cars. Using STAR-CCM+ CFD simulations, twelve wing configurations were tested with variations in endplate angle and wingspan. The goal was to balance aerodynamic efficiency (downforce and drag) for the lead car while minimizing wake disruption for the following car.

This research identifies that small geometric changes significantly alter vortex strength, wake diffusion, and total-pressure recovery, affecting both performance and overtaking potential. The findings also extend to wider engineering applications such as vehicle platooning, multi-aircraft formations, and energy-efficient transport, where wake interference impacts performance.

To achieve accuracy, the study employed Computational Fluid Dynamics (CFD) based on the Navier–Stokes and Reynolds-Averaged Navier–Stokes (RANS) equations, incorporating k–ε and SST k–ω turbulence models. A mesh independence analysis determined that increasing mesh cell counts beyond 1.2×10? provided negligible improvement in accuracy, optimizing the balance between precision and computational cost.

A wind tunnel model (15 m × 2.4 m × 2.2 m) was simulated to replicate real racing conditions, with a free-stream velocity of 50 m/s and a 5 m gap between the leading and trailing rear wings. Results confirmed that turbulent wakes from the lead car significantly reduce the downforce of the trailing car (up to 40%), validating FIA concerns about dirty air effects.

The polyhedral mesh technique was chosen for its accuracy and efficiency in complex geometries. The k–ε model yielded the most reliable lift and drag coefficient predictions. The overall CFD framework ensured mesh refinement validation and transient turbulence modeling, increasing confidence in results.

Ultimately, the research highlights how optimizing rear-wing geometry can enhance aerodynamic efficiency, reduce wake turbulence, and improve close racing conditions. These insights are crucial for co-designing future Formula 1 cars that balance performance and raceability, and for broader aerodynamic innovations across transportation systems.

Conclusion

The findings aimed to investigate the aerodynamic performances of various topologies focusing on identifying and highlighting the reduction of dirty air for the best use cases. It also investigated how rear-wing topology influences wake quality and dirty-air effects in a tandem-wing configuration. The analysis quantified dirty air using the follower’s load change relative to the baseline topology (A), supported by total-pressure and velocity-field diagnostics. The results demonstrate that no single configuration simultaneously optimises aerodynamic performance for both the leading and trailing wings. For the leading wing, the 8° endplate geometries—Topology B (100%, 8°) and Topology J (80%, 8°)—produced the highest downforce increases of 17.31% and 21.33%, accompanied by induced-drag penalties of 12.92% and 16.93%, respectively. These strong suction fields intensified the wake’s total-pressure deficit and downwash, resulting in increased dirty air for a following element. Correspondingly, the trailing wing experienced downforce losses of 30.8% (B) and 24.3% (J) relative to the baseline, confirming the detrimental wake impact of these leader-optimised designs. In contrast, Topology C (100%, 16°) and Topology E (90%, 0°) demonstrated measurable dirty-air reduction. C narrowed the wake corridor and lifted the lowest-pressure deficit, enabling the follower to recover 7.38% additional downforce with a minor drag reduction of 1.939 N (~0.69%). E, through geometric downsizing, weakened tip-vortex intensity and induced downwash, achieving a drag reduction of 19.065 N (~6.76%) with a 2.62% downforce increase. Relative to A, these represent ?7.38% and ?2.62% dirty-air changes, translating to 36–55% recovery of the follower’s lost load compared with B/J. From the results it was clear that there was no linear relationship between decreasing the overall size and increasing the endplate angle simultaneously, with aerodynamic improvements. However, it can be concluded that endplate angle of 8 degrees for leading wing topology was found to be the best performing among all the use cases. Similarly, for the trailing wing topologies, Topology E was the best performing of all. From the data investigated it was also concludedthat Topology E will perform betterwhen it retains full reference size or as close to that size as possible. Overall, B and J maximise leading-wing performance but intensify dirty air, whereas C and E yield cleaner wakes and greater system-level efficiency. The baseline A remains aerodynamically neutral and stable. These findings meet the study objective of identifying geometries that mitigate dirty-air effects while preserving aerodynamic balance across both elements. There was no rear wing topology which enhanced the aerodynamic performance of both leading and trailing wings at a similar level. This indicates that one can only improve the performance of leading wing easily, whereas trailing wing optimization needs more elements to support and enhance its performance. Dirty air reduction was observed with B and J leading wing topologies, but it did not show improvement on trailing wings mainly due to it retaining the same reference wing topology. It is clear from the data obtained that reducing size and increasing end plate angles help in improving performance. The challenging part is to find which combination works for both wings.Future work should focus on (i) mapping total-pressure deficit and turbulence-kinetic-energy decay across the inter-element region, (ii) co-optimising leading and trailing geometries with dirty-air reduction as a design objective, and (iii) exploring the 8–16° endplate-angle range at 90–100% scale to locate an equilibrium configuration capable of enhancing both leader and follower performance.

References

[1] J. Katz, “Aerodynamics of Race Cars,” Annu. Rev. Fluid Mech., 2006. doi: 10.1146/annurev.fluid. (Use this as the core race-car aero text for forces, RANS/Boussinesq notes. If your copy is the 2005 preprint, keep year as in your record.) [2] A. Buljac, H. Kozmar, and I. Džijan, “Aerodynamic performance of the underbody and wings of an open-wheel race car,” Transactions of Famena, vol. 40, no. 2, pp. 19–34, 2016. doi: 10.21278/TOF.40202. [3] M. D. Soso and P. A. Wilson, “Aerodynamics of a wing in ground effect in generic racing car wake flows,” Proc. IMechE Part D: J. Automobile Engineering, vol. 220, no. 1, pp. 1–13, 2006. doi: 10.1243/095440705X69632. [4] J. Newbon and D. Sims-Williams, “Aerodynamic Analysis of Grand Prix Cars Operating in Wake Flows,” SAE Technical Paper 2017-01-1546, 2017. doi: 10.4271/2017-01-1546. [5] M. Wilson, R. Dominy, and A. Straker, “The Aerodynamic Characteristics of a Race Car Wing Operating in a Wake,” SAE, vol. 1, no. 1, 2008. [6] J. P. Holman and J. Lloyd, McGraw-Hill Series in Mechanical Engineering (fundamentals used for CFD pipeline and N–S statements). [7] M. A. R. V. Rayen, “Numerical Evaluation of Underbody Aerodynamics in F1 2022 Car,” ResearchGate, 2019. [8] T. Hughes, “CFD Study of Flow over a Simplified Car (Ahmed Body) Using Different Turbulence Models,” ResearchGate, 2018. [9] D. G. Karkoulias, E. D. Tzoganis, A. G. Panagiotopoulos, S. G. D. Acheimastos, and D. P. Margaris, “Computational Fluid Dynamics Study of Wing in Air Flow and Air–Solid Flow Using Three Different Meshing Techniques…,” Computation, vol. 10, no. 3, 2022. doi: 10.3390/computation10030034. [10] G. M. Le Good, J. P. Howell, M. A. Passmore, and A. Cogotti, “A comparison of on-road aerodynamic drag measurements with wind tunnel data from Pininfarina and MIRA,” SAE Technical Paper 980394, 1998. doi: 10.4271/980394. [11] Z. Wang, X. Hu, Z. Hui, G. Yu, W. Lan, and F. Liu, “Aerodynamic characteristics of MIRA automobile model based on fluid-structure coupling,” AIP Advances, vol. 12, no. 3, 2022. doi: 10.1063/5.0083618. [12] J. Fields, “How Advancements in Aerodynamics Improves the Performance of Formula 1 Racecars,” Union Digital Works, 2015. [13] P. Pansare, S. Chawla, P. B. Pansare, E. K. Mishra, G. Brahma, and R. Kumar, “Application of Mosaic Mesh on Store Separation Modelling and its Validation with Experimental Data,” ResearchGate. [14] D. Gogel, “The Effects of End Plates on Downforce in Yaw,” SAE Technical Paper 2006-01-3647, 2006. [15] M. S. Selig, “An Angle of Attack Correction Scheme for the Design of Low Aspect Ratio Wings With Endplates,” SAE Technical Paper 2002-01-3292, 2002. [16] M. Basso, C. Cravero, and D. Marsano, “Aerodynamic effect of the gurney flap on the front wing of a F1 car…,” Energies, vol. 14, no. 8, 2021. doi: 10.3390/en14082059. [17] S. Jain, N. Sitaram, and S. Krishnaswamy, “Effect of Reynolds number on aerodynamics of airfoil with gurney flap,” Int. J. Rotating Machinery, 2015. doi: 10.1155/2015/628632.

Copyright

Copyright © 2025 Abhinav Sharma, RishI Gupta, Diksha Khuas. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.