Home

Aerodynamics & Airspeed

There is an abundant professional literature that describes the theory, experiments, and physical and numerical models that have been developed and applied to explain the wake generated by the main rotor and its interactions with the helicopter frame while hovering and in forward flight (e.g., Leishman 2006). Yet partly to dispel the “downwash” misconception, and mostly to identify the range of airspeeds that can provide the best results for measuring fluxes and aerosols (which is essential for research mission planning), we used an analytical model, numerical simulations, and observations collected on board the HOP. These studies are summarized below.

Analytical Study

Rotor performance inflight was first derived and explained by Glauert (1935) based on the analysis of marine propellers proposed by Rankine (1865) that was further developed by Froude (1878) and Froude (1889). It is often referred to as the “Rankine-Froude Momentum Theory.” It is thoroughly described in most introductory textbooks on helicopter aerodynamics (e.g., Leishman, 2006) and a simplification of this flow model was adopted here. It is described in Avissar et al. 2008a.

The figure below shows the magnitude of the airflow velocity through the rotor, and the resultant airflow velocity and the angle of the rotor wake obtained with this model at various airspeeds, from hover to the maximum cruising speed of ~60 m/s. It is interesting to note that the induced-air velocity decreases rapidly with airspeed and, as a result, the wake angle switches from vertical at hover to about 68o at an airspeed of 15 m/s. The angle between the tip of the blade (when aligned with the front of the helicopter) and the nose of the helicopter is about 57o, which is cleared of the rotor wake at an airspeed of about 10 m/s. This very simple analysis is clearly supported by the observations of Leishman and Bagai (1991).

Numerical Study

FLUENT (www.fluent.com), a state-of-the-art commercial computational fluid dynamics (CFD) software, was used to simulate the Jet Ranger in flight at different airspeeds. The figures bellow illustrate some of the simulation results, and a detailed description of their setup and analysis is provided in Avissar et al (2008b). Initially, the streamlines are horizontal and the background airspeed is constant in space and time. Therefore, any impact from the helicopter on the airflow is seen on these graphs as a departure of the streamlines from horizontal and/or a change of color (shown below).vortex simulated with FLUENT It is interesting to note that, concerning the main-rotor wake position at different airspeeds, there is no conceptual difference between these results and those obtained with the quite simple analytical study discussed in the previous section. This emphasizes the robustness of the assumptions and simplifications made in our analytical study. However, an important result of the CFD simulations shows that as the helicopter flies faster and faster, a “pocket” of compressed air develops and grows in front of it, creating another zone of air disturbance that is independent of the main rotor. This additional disturbance is similar to that observed in front of airplanes, and it is affected by the aerodynamic shape of the aircraft structure as well as its airspeed. This is well simulated with the CFD but ignored in the analytical study. Based on these results, it appears that an airspeed of at least 15 m/s is needed to clear the Jet Ranger nose from its main-rotor wake, but to avoid strong airframe-induced disturbances, it seems best to maintain air speeds below 50 m/s.

 


Observational Study

TKEThe mount used to attach the sensors in front of the HOP was partly designed based on the above results. Another consideration in its design was vibration reduction. To calibrate these sensors, evaluate their performance in flight, and provide additional insights on the operating range of the HOP, we performed two low-level flights (i.e., ~15 m ASL) at various airspeeds at the Outer Banks of North Carolina. The marine boundary layer (MBL) is typically more homogeneous than the continental one, and we carried out our observations there to minimize the change of turbulence during the flights, which each lasted about 45 minutes. We selected a day with easterly winds (i.e., from the sea) and flew about 200 m offshore to avoid land effects. For each flight, we maintained a constant airspeed of 20, 25, 30, 35, 40, 45, 50 and 55 m/s for a period of about 5 minutes each.

The calibration of the sensors and the procedure used to detrend the data and to eliminate the helicopter vibration from them is discussed in Holder and Avissar (2008). Here, the focus is on the operating range of the HOP and, therefore, we are mostly concerned with the relative values of the atmospheric variables at different airspeeds. The figure at left indicates that no main-rotor impact can be detected on the turbulence kinetic energy (TKE), CO2 flux, and sensible and latent heat fluxes at airspeeds above 15 m/s. We do notice a higher TKE at the lowest airspeed, which does not seem to affect the calculation of the vertical fluxes. This is probably due to lateral movements of the helicopter, which is more sensitive to wind gusts at low speed, but is unlikely associated with the rotor wake given the lack of impact on the vertical velocity used to calculate the fluxes. We plan to perform an additional series of low-speed flights (i.e, 5 – 25 m/s) and study this issue in more details, especially if we need to prepare for a mission that needs to be flown at the lowest possible airspeed.

Wave numbersThere seems to be a small decreasing trend in TKE and fluxes as airspeed increases above ~45 m/s. However, due to their small value in the MBL, this trend is not very obvious. While from a practical point of view, it is enough to conclude that the operating range of the HOP is optimal from 15 to 45m/s, to explore this issue in more detail, we produced the spectra of the vertical wind component (w) at various airspeeds (see figure at right). As expected, the de-noised spectra gradually fall off at lower wave numbers as airspeed increases, so that the size of the smallest eddies that are adequately sampled increases. When we will have an opportunity to fly in the Western USA over a homogeneous, hot land surface, which produces very-high turbulent sensible heat fluxes and a well developed CBL, we intend to perform additional flights to continue investigating this issue.

References

  • Avissar, R., H. E. Holder, P. Canning, K. Prince, N. Matayoshi, R. L. Walko, and K. M. Johnson, 2008a. The Duke University Helicopter Observation Platform. Bulletin of the American Meteorological Society, submitted.
  • Avissar, R., N. Abehserra, P. Canning and H.E. Holder, 2008b. Simulations of the operating range of the Duke University Helicopter Observation Platform. J. Atmos. Ocean. Tech., in preparation.
  • Froude, R.E., 1889. On the part played in propulsion by differences of fluid pressure. Transactions of the Institute of Naval Architects, 30, p. 390.
  • Froude, W., 1878. On the elementary relation between pitch, slip and propulsive efficiency. Transactions of the Institute of Naval Architects, 19, pp. 47-57.
  • Glauert, H., 1935. Airplane Propellers, In Division L of Aerodynamic Theory, ed. W.F. Durand, Springer Verlag, Berlin, Germany. Reprinter by Peter Smith, Glouster, MA, 1976.
  • Holder, H.E. and R. Avissar, 2008. Measurements of turbulent fluxes in the atmospheric boundary layer with the Duke University Helicopter Observation Platform. J. Atmos. Ocean. Tech., in preparation.
  • Leishman, J.G., 2006. Principles of Helicopter Aerodynamics. Cambridge Aerospace Series. W. Shyy and M.J. Rycroft, Editors, Second Edition, 826 pages.
  • Leishman, J.G. and A. Bagai, 1991. Rotor Wake Visualization in Low-Speed Forward Flight, AIAA Paper 91-3232, 9th AIAA Applied Aerodynamics Conference, Baltimore, MD, Sept 23-25, 1991.
  • Rankine, W.J.M., 1865. On the mechanical principles of the action of propellers. Transactions of the Institute of Naval Architects, 6, pp. 13-39.