Aerodynamic design I

As it has been seen before, during the description of the propulsion system, is absolutely necessary to know some aerodynamic characteristics in order to determine the performance of the airplane and therefore to check if the aircraft meets the design requirements. To facilitate the task, the plane has been divided into the following parts: wing, fuselage, tail and control surfaces. The first two are discussed in this article, the others will be discussed along with the stability of the plane.

In this design phase, besides the necessary theoretical analysis, two computer tools have been used: Javafoil for 2D airflow study and the airfoil selection and Ansys CFX, for the visualization of 3D airflow over every element as well as to determine the global forces and moments over the airplane.

Most of the aerodynamic design is optimized for cruise flight due to airplane spends most of the time in this situation. 

Wing:

The wing design begins with the selection of the airfoil or airfoils that will be used on its construction. The airfoils chosen must be fitted to flight conditions (Mach and Reynolds numbers), must assure the correct behavior of the boundary layer and must achieve enough lift force at the same time that tries to minimize the associated drag.

From the propulsion system study the cruise speed was stablished between 80 and 120 meters per second at an altitude of around 40000 feet so the Mach number will be under 0.4 in nominal flight conditions. At that altitude and speed with a MAC of 1.35 meters the Reynolds number will be around 5x10⁶. The first study with thin laminar airfoils showed a good performance for low speeds. At that speeds and Reynolds intervals, the boundary layer remains laminar and attached even at 50 percent of the chord.

Some of the airfoils tested:

Its dependence of lift coefficient with attack angle and polar for ideal cases (infinite wing and clean surface conditions) where shown below:

  

The NACA 6 series showed the best results in drag, with and efficiency factor over 50 with NACA standard surface conditions. Besides the good results that shows these thin airfoils, was necessary to increase the thick of the wing to allow enough fuel volume inside the wings and to prevent structural problems. I decided to keep trying with NACA 6 family and after several tests the final airfoil selected was NACA 64-517 (a=0.9).

 Flow field

 Polar

 cl vs attack angle

 Numerical results for Re=5x10⁶

With the airfoil data and an estimation of the maximum weight at cruise (steady level) conditions, the wing surface can be easily calculated. Must be remembered that the airplane will spend most of the flight time in this conditions.

Considering as cruise conditions h=40000ft, V_T=90m/s and MTOW=3800kg, the wing surface needed to maintain a steady level flight can be expressed as a function of the lift coefficient as follows:

A few values to illustrate this relationship:

Cl	0.3	0.45	0.6	0.75	0.9	1.05
Sw (m2)	84.2	56.1	42.1	33.7	28.1	26.5

An early design of the wing attending to the selected airfoil with a slight modification in taper ratio (almost 0.45) and sweep angle is presented below. With this modification is intended to obtain a surface of around 30 square meters, a smoother transition for the installation of the winglet at tip airfoil and a lift force with almost elliptical distribution.

With a surface of 30m² (as shows the above design) the initial lift coefficient that must be achieved at cruise conditions is 0.84 By interpolation in the airfoil data, it means that the angle of attack needed to maintain a steady level flight would be approximately 2.4^o. According the consulted bibliography, this angle is on the normal range for commercial aircrafts at initial cruise conditions. In the same way at the end of the cruise phase the aircraft weight will be of only 2000kg, the lift coefficient needed will fall to 0.46 and the associated angle of attack will be of -0.6^o. This variation of 3 degrees during the cruise has to be accounted in order set an appropriate offset angle between the body and the wing of the aircraft.

Propulsion system

 

As endurance is established as a main requirement, the propulsion system must be one of the first elements to be evaluated. Considering the state of art, a solar powered propeller aircraft would attain the best endurance performance, reaching in some cases more than 30 days of continuous flight. Unfortunately due to structural and weight limitations, these kind of planes allows a very limited payload, so has been discarded its use.

After consider the rest of propulsion systems, the turboprop seems to be the best option to achieve the adequate compromise between endurance, payload and flying altitude.

The weight of a turboprop engine is about half that of a comparable piston engine and have some advantages over a turbojet. For example: The amount of power available for propulsion is largely independent of the forward speed of the aircraft, so that more power is available during the initial stages of the takeoff run. The engine can be run under more efficient and economical conditions at low and medium altitudes, and retains these two qualities at low aircraft speeds. With the use of interconnected engine and propeller controls, the power response to throttle movement is more rapid than that of a turbojet engine and operations can be conducted from shorter runways.

Two commercial turboprop models had been considered for been the power plant of the UAV, PT6A (made by Pratt & Whitney Canada) and TP10 (made by Honeywell). Below tables with its performance are shown using the data available on internet. Despite the usual units in bibliography are horsepower and pounds, I have preferred to use SI units for coherence with the rest of data and calculations.

Model	SP (Kw)	Cp (kg/Kwh)	m* (kg/s)	Length (m)	Diameter (m)	Weight (kg)
PT6A 36A	559,3	0,359	3 - 4,5	1,575	0,483	150
PT6A 66	633,8	0,62		1,778	0,483	213
PT6A 135	559,3	0,585		1,778	0,483	156

Performance data and dimensions of some PT6A mdels

Model	SP (kw)	ESP (Kw)	SFC (kg/Kwh)	m* (kg/s)	Weight (kg)
TPE331-10	670	704	0,35	3,5	190

Performance data of TPE331-10 turbprop

For the Honeywell TPE331-10 also are shown a geometrical sketch and charts of fuel consumption and shaft power depending on the true air speed and altitude.

From these data can be drawn some interesting conclusions:

• Considering a long range cruise speed of 80 m/s (150-160 Knots) at an altitude of 12000m (above 40000 ft.) the approximate fuel consumption is 63 kg/h, which means around 1500 kg of fuel to fulfill the 24 hours goal. This figure is only a rough approximation and aerodynamic data will be necessary to obtain more accurate results, but suggests that an initial consideration of 1800 kg of fuel (1500kg plus 20%) may be appropriate for first calculations.

• With an aerodynamic efficiency factor of 30 for clean configuration (somewhat higher than in commercial jets but far less than in sailplanes) and assuming steady level flight the maximum weight of the airplane can be expressed as:

For the TPE331-10 engine, with a propeller efficiency of 0.8 and supposing long range cruise speed and same altitude, the weight limitation due to maximum shaft power, would be of 6400 kg. Although there are more restrictive factors such as wing loading, 6000 kg for MTOW was one of the first references in order to extrapolate the weight data of the similar aircrafts.

• Besides the relevance of the engine performance over the design also has to be considered its effect on operations. One of the most clear is that limitation in altitude is established by the engine operation limit, at 45000 feet, 5000 feet less than expected. On other hand, to allow an easy comprehension of the range of speeds that are achievable with this engine graphs of power against airspeed are showed below:

 Considerations about general arrangement of the airplane.

Final arrangement of the UAV with a fuselage new design

After speaking, during the preceding entries, in very general terms, it is time to present a more detailed description of the preliminary design solutions, element by element.

Wings:

Straight trapezoidal: Easier construction, good behavior at moderate speeds (up to Mach 0.3). With a taper ratio of approximately 0.44 provides a near elliptical lift distribution, minimizing the induced drag.

High aspect ratio (near 20): Less drag for the same wing surface. Not too high in order to avoid structural problems.

Use of Winglets: Less drag and better aerodynamic efficiency.

Under fuselage construction: Better compatibility with horizontal stabilizer and easier integration of the landing gear system.    

Wing and horizontal stabilizer diagram

Body:

Circular cross sections: Due to the moderate width of the body these kind of sections provide some constructive and operational advantages without compromising the drag.

Front body wider and rounded: More space available for payload. The rounded surfaces allow better performance of optical and radar-based devices.

Rear body adapted for propulsion: The main section width is determined by the engine dimensions. The rear pusher propeller, with its body-centered axis, allows to reduce drag and cause less interferences with the navigation avionics and payload. The air intake is situated over the fuselage providing a cleaner airflow, especially during take-off.

Lateral view of the aircraft

Tail:

H configuration: Trying to avoid interferences between the tail and the rear mounted turboprop. Also allow that longitudinal and lateral stability problems could be solved separately.

Aerodynamic interaction between the wing and tail.

Preliminary design

In the very initial stages of the design of a non-conventional airplane like this, there are a lot of different paths that can be taken. Maybe the first idea that can cross the mind of an enthusiastic engineer is to develop an innovative concept that perfectly suits to performance needs. Of course, that way could be better in order to achieve the requirements but usually implies to work out new technology and for sure this option will increase dramatically the amount of effort, time and money that the project needs.

A more reasonable or conservative approach is make a less risky bet, using some of the design solutions applied to similar airplanes and focusing the work in a suitable arrangement of these solutions in order to meet the desired requirements and finally achieve a complete new design.

This kind of work philosophy was applied to this project and two main types of airplanes were considered for inspiration: Sailplanes and General Atomics’ UAVS.

Due to the need of a high aerodynamic efficiency, sailplanes wings and body were a reliable design reference. On the other hand, General Atomics’ Predator is a fully proven airplane with similar MTOW and NASA’s Altair prototype seems to have very similar performance requirements.

Here are some photos of the planes that were taken as a reference:

Sailplane during flight

NASA's Altair unmmaned aircraft

Predator US military drone

Researching on internet a valuable amount of data about the reference planes like weight, wingspan or even airfoils can be collected.

At this point the project have the mission requirements defined and a few models to extrapolate data. Building on the simple work philosophy which it was spoken above, just missing enlist the help of a design handbook to begin developing the project. Easy, right?

Well, at the end was not so easy, but after many hours of reading datasheets, making calculations and working with CAD programs, a first design was released.

This is only a basic model with the aircraft elements roughly sized where all the ideas and technical solutions are put together.

Of course this design will need to change in many aspects, because every time that is make a new calculation or a deeper analysis, the results will show that the fuselage must be wider or maybe that the tail surface is not enough to ensure stability. But even if a lot of changes must be implemented at least a core of this design will remain constant due to the general arrangement of the plane has been established and in general terms the associated technical solutions that have been adopted are still valid.

A new project

After finishing my master thesis I really thought that would be a long time before returning to undertake such a long and tiring work, but only a few months later I was already involved in a challenge that will need even more time and effort.  So from now I will publish on this blog the different stages that I am facing in my attempt to develop a realistic design of a long-endurance drone.

Why an unmanned plane? There are many reasons but I believe that all began with a simple idea, something as simple as trying to find a more dynamic and affordable alternative to do some of the tasks that nowadays are being carried out by low earth orbit satellites.

The unmanned vehicles are a reality, a technology that that has been showing spectacular results in the last years. Some of its advantages in civil applications include improves in weight and distribution due to the suppression of the life support systems, avoid human casualties in dangerous missions as fire extinction and even may have lower operational costs than conventional planes.

 Leveraging these advantages, the real objective is to create with a few airplanes a multi-role platform able to provide internet services to a remote place during one week and if it is necessary, the day after, collaborate in rescue tasks only by changing the payload.

Having this in mind, the main features that I believe must accomplish the drone plane in design are: 

-          

•        To have an endurance of at least 24 hours.
•           Ensure a fully autonomous flight (no necessarily a man in the loop).
•           Be able to carry an interchangeable payload of at least 300 kilograms.

Other not mandatory but desirable requisite to be achieved in the final design would be a ceiling around 50000 feet to obtain a good coverage.