IJUSEng - Aerodynamic Design of Delta UAV
IJUSEng - Aerodynamic Design of Delta UAV
Technical Note
IJUSEng – 2014, Vol. 2, No. S2, 1-15
http://dx.doi.org/10.14323/ijuseng.2014.4
Alexander S. Goodman
Marques Aviation Ltd, USA
1. INTRODUCTION
This paper describes the conceptual aerodynamic design of a variant of a delta blended
wing-body tailless mid-range unmanned aerial vehicle (UAV) with intelligence, surveillance
and reconnaissance (ISR) and possible strike capabilities. The tailless configuration is
desirable for a number of reasons. With no vertical control surfaces, or at least no large
vertical surfaces, the aircraft has minimal radar signature.[1] This configuration can produce
lower drag than conventional aircraft and greater performance at high angles of attack (α).
Also, with the lack of inherent flight stability, the aircraft can perform combat manoeuvres.[2,3]
With the limitations in technology and materials in the past, the tailless design layout was not
as efficient as the conventional layout and was also very unstable. With advancements in
both technology and materials a tailless delta configuration can be designed to yield greater
performance than conventional-style aircraft.[4] The performance of tailless aircraft is
expected to improve with further experimentation and analyses. This paper addresses the
following topics of aerodynamic design: airfoil selection, basic aerodynamic modelling for
steady state stability, planform shape, planform geometric alterations, control surfaces,
adaptive wing technology, and aeroelasticy.
2. AIRFOIL SELECTION
Traditional cambered airfoils produce a negative pitching moment (Cm), nose-down effect, on
the airfoil. This is counteracted through the empennage by the horizontal stabilators. In a
tailless delta-wing type aircraft, careful selection of the airfoils is essential, since Cm strongly
contributes to the aerodynamic longitudinal stability of the aircraft.[5] The Cm is measured
around the aerodynamic centre (a.c.). Both Cm and a.c. are discussed in further detail in the
next section dealing with aerodynamic modelling. With no tail for longitudinal stability, the
airfoils selected should have low or zero Cm. Instead of using a symmetric airfoil, which has
zero Cm at zero α, a suitable solution is to choose a reflexed airfoil.[4] The following presents
an analysis on reflex airfoils. The camber shape on an airfoil is closely related to Cm. The aft
part of the camber line has a great influence on Cm. Because of this influence, the aft end of
the camber line is reflexed. By increasing the degree of reflex, the Cm vs. α curve shifts
upward (positive). The drag polar is also affected and shifts down. This produces an
undesired reduction in the maximum lift coefficient, but can be fixed by creating more
camber. However, greater caber increases the negative nose down Cm. In aerodynamics this
is usually the case, improvement of one aspect negatively affects other aspects of the
aerodynamic design. Another option is to change the location of the maximum camber. The
position of the maximum camber has a small influence on the drag polar, but can greatly
impact the pitching moment.[6] If the maximum camber position is shifted backwards, the Cm
increases in the negative direction (nose-down). The maximum camber position should,
therefore, be towards the leading edge (LE) of the airfoil for a delta planform aircraft. This is
illustrated in Figs. 1 and 2, where three different airfoils are compared.
A reflex camber airfoil is used for the outboard portion of the wing. The NACA 5 digit series
airfoils are known to produce low Cm. Specifically, the NACA 22112 airfoil was selected (Fig.
4). The NACA 22112 airfoil was then compared to a similar airfoil, NACA 23112, which
differs from the NACA 22112 airfoil only by the farther aft position of the maximum camber
location.[5] These airfoils were compared at the two Re of 0.5 x 106 and 1 x 106. The results
are shown in Table 1. Plots of airfoil aerodynamic coefficients vs. α can be found in Figs.
A.1-A.4 in the Appendix. It can be seen from the Figures that as the maximum camber is
pushed aft, the maximum lift decreases and the nose down Cm increases. The combination
of these two airfoils may balance each other, depending on the spanwise distance
distribution, and could create a slight washout.
Airfoil and Re
3. AERODYNAMIC MODELLING
One of the most important parameters of an aircraft is the a.c. The a.c. is the point on the
wing mean aerodynamic chord where the variation of Cm with α is zero. Finding the a.c. on
the wing planform is very important, however the influence of the fuselage must be
considered. When a fuselage is added to a wing, the a.c. of the wing-fuselage combination
shifts forward compared to that for a wing. This means that the fuselage will add a positive
Cm with each increase of α. This is called Munk effect.[9] The design of the proposed aircraft,
however, resembles that of a blended wing-body similar to that of a flying wing. Thus, the
addition of the fuselage will not play as big a role on the Cm as a conventional aircraft, but
still cannot be ignored.[4] In order to know how the aircraft will perform, modelling of the
aerodynamic and thrust forces and moments must be determined for steady state and
perturbed state. There are two ways to determine these forces and moments. These are
experimental methods, such as flight or wind tunnel tests, and computational and/or
empirical methods. The following presents basic mathematical models to show how to
determine the aerodynamic forces, thrust forces, and moments. For the case of tailless
aircraft conventional modelling is not appropriate, rather the wing-body effects need to be
included. The steady state aerodynamic forces and moments are discussed below. They are
broken down into longitudinal variables and lateral-directional variables. An in-depth analysis
of these models is beyond the scope of this paper. Roskam[9] and Napolitano[10] discuss this
topic in greater detail. Thrust and perturbated state forces and moments are not covered
below. Perturbated state modelling is complex and it is a subject beyond the basic
conceptual design presented in this paper. The following modelling data and equations were
taken from Roskam[9] and Napolitano.[10]
3.8. Side-Force
The equation for is ; where, = side-force coefficient. The steady state
airplane aerodynamic is dependent on the sideslip angle, α, Mach number, Re, and the
deflection of lateral and directional control surfaces. For an in-depth analysis, four
aerodynamic coefficients are considered important. According to Napolitano[10], these
coefficients are referenced as the four pillars because they need to fit together in order for
the aircraft to fly adequately. The four coefficients are:
4. WING PLANFORM
The wing planform for the proposed UAV is a variant of the delta configuration incorporating
a blended wing-body without a tail. Examples of different delta shape planforms can be seen
in Fig. 5. This planform shape was chosen to reduce radar signature for ISR and possible
strike missions. The delta configuration is very common in high speed fighter jets because
the shape allows the aircraft to cruise at very high subsonic speeds and achieve high α
without stalling. This is accomplished due to the leading edge vortices that flow over the
wing at high α.
In a tailless configuration, the wing span is extended farther back than the base of the
aircraft. The airframe shape resembles that of a notched delta. Examples of similar wing
planforms can be observed in the Northrop Grumman Bat and the Lockheed Martin RQ-170
Sentinel (Figs. 6 and 7). The notched delta planform balances out the centre of gravity along
with the desired control surfaces to add greater manoeuvrability and control.[11]
There are two important vortex patterns that form over delta wings, a primary vortex and a
secondary vortex.[6] The primary vortex is formed by the sharpness of the leading edge
causing the flow shear to separate. The primary vortex consists of a pair of strong leading
edge vortices characterised by low local pressure.[12] With the pressure drop caused by the
primary vortices the lift is enhanced allowing the delta configuration to operate at higher α
than conventional planforms. The formation of these primary vortices usually appears after a
certain positive α is reached and becomes stronger with higher α, thus producing more lift.
Eventually, the stall angle is reached caused by vortex break down.[12] An in-depth visual
description is displayed in Fig. 8.
Fig. 8: Flow over a delta wing planform (based upon Anderson [6])
An important characteristic of delta wings is the leading edge radius. Depending on the
mission, the radius needs to be tailored. For completely subsonic missions and hypersonic
missions, such as in the Space Shuttle, a larger rounded edge should be used. The
aerodynamics behind these two missions is different. The hypersonic re-entry of the Space
Shuttle allows extreme high temperatures without damaging the wing. The large rounded
edge is also used to create a large bow shock to slow the aircraft down.[6] The larger
rounded edge for low subsonic flight missions is used to create higher lift to accompany the
low speeds, as well as keeping the flow attached for longer.[5]
5.2. Sweep
The proposed tailles UAV is modeled after a delta wing. Therefore, wing sweep needs to be
implemented. It has been discussed earlier that the delta type planform experiences
additional lift through the vortical flow over the wing. The amount of wing sweep can greatly
alter the aerodynamic characteristics of the aircraft and should be determined. The aircraft
can have varying leading edge and trailing edge sweeps. The α seen by each wing segment
increases towards a swept wing tip.[1] This causes the wingtips to stall first, which needs to
be avoided. The wing sweep affects the aerodynamic centre, lift curve slope, and zero-lift
slope along with many other individual parameters, such as dihedral effect.[9] Since the
sweep angle is such a critical parameter, it should be carefully analyzed and tested to
determine which sweep angle best fits the desired performance of the UAV.
6. CONTROL SURFACES
UAVs can have numerous types of devices for stability and control. With a lack of an
empennage, the aircraft requires non-conventional control surfaces and/or have them placed
in non-conventional positions in order to achieve stability and control.[1] The use of trailing-
edge high-lift devices should be avoided. These devices create large pitching moments
causing longitudinal instability from which the aircraft may never recover.[13] With the location
of the wingtips protruding behind the trailing edge of the aircraft, the elevons can function as
elevators and ailerons for pitch control as well as roll control.[11] The wing can be equipped
with either large elevons to compensate for the small moment arm or have numerous
smaller elevons.[3] Using several smaller elevons could prove beneficial for small
adjustments in flight attitude because such configuration creates less drag and more easily
dampens the aircraft’s oscillations. Using numerous smaller elevons instead of one large
elevon can also lessen the effect of decambering along the span when the control surfaces
are deflected. Many aircraft use wingtip devices such as winglets. Winglets drain the energy
from the tip vortices and change the lift distribution over the wing (Fig. 9).[2]
Since the proposed UAV does not use a vertical tail, winglets could serve a dual purpose if
designed properly. The winglets can help lower the strength of the tip vortices and act as
rudders for lateral/directional stability and control. The rudders at the wingtips will also take
advantage of the longer moment arm, thus needing less deflection for a desired attitude
adjustment.[11] A Whitcomb winglet might prove to be more beneficial to this UAV than
conventional winglet designs (Fig. 9). Whitcomb winglets decrease the strength of the tip
vortices and turn that energy into forward propulsion.[3] The use of winglets causes less air
disturbances produced by the wing and reduce Doppler radar detection, but at the same
time the vertical structures will cause larger cross-sectional radar detection. As stronger,
lighter and smarter materials are developed more effective control surfaces may be
incorporated into the design of the delta aircraft.
Adaptive wing technology can greatly improve an aircraft’s performance without the
bulkiness and complexity of conventional devices. There are two types of control technology,
active and passive control. Active control requires additional energy and/or the presence of
feedback, while passive control is self-regulatory. Some methods used in adaptive flow
control are: boundary layer suction, perturbation control, laminar control, shock control,
turbulence control, separation control, load control, variable camber, and changes in the
structure through aeroelasticity and damping.[15] These can further be broken down into
geometric and pneumatic devices. A few of these devices are: vortex generators, Gurney
flaps, reversed-flow flaps, ribblets, leading edge suction systems, divergent trailing edges,
and zero-mass jets.
eventually damaging the flap.[15] A possible fix is to use an active reverse flow flap instead of
the previously discussed passive, freely movable, flap.
8. CONCLUSION
The design of an aircraft is a very difficult and complex process. It does not matter if the
aircraft is manned, unmanned, small or large; each have their own difficulties. Tailless delta
aircraft have been around for decades, but have recently seen a rise in production. This can
be attributed to the emergence of more advanced materials and technologies that enhance
aircraft performance. It was shown that there are many trade-offs when designing the wing
planform. When one aspect of the design is increased or optimized then at least one, most
likely multiple, aspects or parameters will be decreased. This paper helps create an outline
of a conceptual design of a tailless delta-like UAV, but the analysis presented is not sufficient
for a complete wing design. Many important factors were left out such as materials,
structures, and the propulsion unit. The aerodynamic modelling section was briefly
described, however a detailed analysis of the aerodynamics and thrust forces and moments
need to be conducted in order to create the needed complex flight control system. The flight
control system is the brains of the aircraft and with a subpar system a tailless delta aircraft
will not be stable and possibly not be capable to complete the desired mission. As more
aircraft of this design are developed, the aerodynamic performance will become greater and
a detailed design layout will become just as generalized as a conventional tube and wing
aircraft.
Acknowledgements
The author wishes to thank Marques Aviation Ltd for their support during this project.
9. REFERENCES
1. Friedl J. (2001). Aerodynamic design of a tailless aeroplane. ACTA Polytechnica.
Journal of Advanced Engineering. 41(4/5): 79-84.
2. Loth JL. (2007). Engineering approach to aerodynamics and aircraft performance. (2nd
ed.). McGraw-Hill. New York.
3. Marqués P. (2013). Flight stability and control of tailless lambda unmanned aircraft.
International Journal of Unmanned Systems Engineering. 1(S-2): 1-4.
4. Prisacariu V, Boscoianu M and Cîrciu I. (2013). Design and construction a flying wing
unmanned aerial vehicles. Proceedings of the International Conference on Military
Technologies. 22nd – 23th May. Brno. Czech Republic.
5. Jiangtao S, Hao Z and Junqiang B. (2006). Airfoil design of tailless unmanned air
vehicle (UAV). Proceedings of the 25th International Congress of the Aeronautical
Sciences. 3th - 8th September. Hamburg. Germany. Pp. 1-5.
6. Anderson JD. (2010). Fundamentals of aerodynamics. (5th ed.). McGraw-Hill. New-
York.
7. Aerodesign. (2001). Airfoil database for tailless and flying wings.
(http://www.aerodesign.de/english/profile/profile_s.htm#hs522)
8. Airfoil Tools. (2014). (http://airfoiltools.com/airfoil/details?airfoil=naca22112-jf)
9. Roskam J. (2007). Airplane flight dynamics and automatic flight controls. Design,
Analysis and Research Corporation. Lawrence. USA.
10. Napolitano M. (2011). Aircraft dynamics: From modeling to simulation. John Wiley &
Sons, Inc. United States.
11. Donlan CJ. (1944). An interim report on the stability and control of tailless airplanes.
National Advisory Committee for Aeronautics. 796: 419-435.
12. McLain BK. (2009). Steady and unsteady aerodynamic flow studies over a 1303 UCAV
configuration. PhD thesis. Navy Post Graduate School. Monterey. United States.
13. Loftin LK. (1985). Technology of the jet airplane. Wings and configurations for high-
speed. In Loftin LK. (Ed.). Flight. Question for performance: The evolution of modern
aircraft. NASA Scientific and Technical Information Branch. Washington. D.C. Pp. 180–
195.
14. Weisshaar TA. (2010). Static and classical dynamic aeroelasticity. Vol. 3 Structural
Mechanics. Section 3.2.2. Encyclopedia of Aerospace Engineering. John Wiley. London.
15. Stanewsky E. (2001). Adaptive wing and flow control technology. Progress in
Aerospace Sciences. 37(7): 583-667.
10. NOTATION
a.c. Aerodynamic centre
Span
Mean geometric chord
Drag coefficient
Rolling moment coefficient
Dihedral effect
Lift coefficient
Change in aircraft lift coefficient with α
Pitching moment coefficient
Change in aircraft pitching moment coefficient with α
Yawing moment coefficient
Weathercock effect
Side-force coefficient
Drag force
Side/lateral force
Drag force
Lift force
ISR Intelligence, surveillance and reconnaissance
Lift force
Rolling moment
Pitching moment
Yawing moment
Dynamic pressure
Re Reynolds number
S Planform area
α Angle of attack
11. APPENDIX
Fig. A.1: NACA 22122 airfoil performance at Re = 5 x 105 (Airfoil Tools [8])
Fig. A.2: NACA 22112 airfoil performance at Re = 1 x 106 (Airfoil Tools [8])
Fig. A.3: NACA 23112 airfoil performance at Re = 5 x 105 (Airfoil Tools [8])
Fig. A.4: NACA 23112 airfoil performance at Re = 1 x 106 (Airfoil Tools [8])
Copyright of IJUSEng is the property of Marques Engineering Ltd and its content may not be copied or emailed to multiple sites
or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email
articles for individual use.