Resulting Design: Decode-1

As part of the DECODE program mentioned in the introduction, we designed several aircraft adopting the principles set out in this book. The first of these, Decode-1, was a simple single-engine pusher aircraft that we used to gain information on performance and weights for subsequent design work. A pusher design was selected so as to give an uninterrupted view forward for the payload, but it does mean the fuselage length has to be sufficient to allow the payload mass to balance out the rear-mounted engine mass so as to yield an acceptable center of gravity. The aircraft was also designed to allow a range of different experimental wings to be trialed and was sized so as to fit inside our largest wind tunnel without significant modification. Figures 11.1 and 11.2 show this aircraft with conventional wings attached.

The basic design brief for the aircraft is given in Table 11.7. If this brief, together with the data from Tables 11.1-11.3, is fed into our spreadsheet system, which follows the algorithm previously outlined, the design details given in Tables 11.8 and 11.9 result; see also Figures 11.3-11.5, which show the inputs, results, and simplified layout sketches we create inside the spreadsheet during design studies^[1]. We find that seeing the aircraft in planform and side view in this way is very helpful when making design judgments early on in the design process; it is immediately apparent how similar this is to the layout of the final aircraft.

Note that three of the constraints set out in Table 11.2 are actively limiting the design: the peak wing C_l/C_d at cruise speed is 16; the allowable rotation at takeoff before grounding happens is 17°; and the static margin (expressed in fractions of main wing chord) is 0.1. Of these, it is the C_l/C_d value that is most fundamental aerodynamically, the other two largely dictating the fuselage and tail-boom length. The maximum value of C_l/C_d is set by the wing technology being deployed and depends on the detailed choice of planform, sections, twist,

Figure 11.1 Decode-1 in the R.J. Mitchell wind tunnel with wheels and wing tips removed and electric motor for propeller drive.

Figure 11.2 Decode-1 in flight with nose camera fitted.

Table 11.7 Design brief for Decode-1.

Table 11.8 Resulting concept design from spreadsheet analysis for Decode-1.

The first five entries here are manipulated by the solver tool within the spreadsheet to maximize the range.

Table 11.9 Design geometry from spreadsheet analysis for Decode-1 (in units of mm and to be read in conjunction with Tables 11.3 and 11.8).

and wing tip treatment. The value of 16 used here is intimately related to the wing loading discussed in the previous chapter, and for operations at sea level and 30 m/s gives a wing loading of 15.5 kg/m². The value of C_l/C_d used is something we know we can achieve in practice using the construction methods we have adopted; higher values are possible but these tend to make the wings more difficult and expensive to make. For example, we chose to taper our wings linearly and avoid twist when working with hot-wire-cut foam; twisted or elliptic planform wings would give better control of induced drag, but instead we often make use of quite sophisticated 3D-printed wing tips to limit induced drag, see again Figure 11.2.

Once one is happy that the concept is sufficiently well developed to warrant further effort, the basic geometry details can then be used to generate a more realistic and fully three-dimensional outer envelope for the aircraft. To do this, we use the AirCONICS suite of programs,^[2] which leads to the design shown in Figure 11.6. This can then be used for further analysis or to start the process of building a detailed CAD model suitable for generating manufacturing drawings. Note that the AirCONICS geometry includes wing taper, an aerodynamically shaped circular fuselage, faired in-tail booms, and slab-sided undercarriage legs. These are just working assumptions at this stage, but the resulting model is sufficiently detailed for reasonable CFD and FEA analysis. No attempt at this stage has been made to add control surfaces to the wings, tail, or fins, so any CFD results would be solely for the cruise configuration. These can be added later using AirCONICS, as will be seen in Chapter 13. The final design shown in Figure 11.2 has greater wing taper, wing tips, and a triangular section fuselage, but otherwise is broadly similar to this initial AirCONICS model.

• [1] Note the highly structured spreadsheet layout we adopt, which also always show the formulae being used alongsidethe results calculated - the geometry page is additionally structured to allow a simple cut-and-paste into the pythoncode we use to invoke the AirCONICS CAD package.
• [2] https://aircraftgeometrycodes.wordpress.com/airconics/.