Propeller for High-Advance-Ratio Interceptor UAV Propulsion — A Trimmed and Re-Twisted DA4052

1. Motivation

Small-scale propeller research is concentrated at low advance ratio. Published databases, including the campaign from which the parent blade is drawn, characterise rotors at operating points typical of hover, climb and modest forward flight. Interceptor-class UAVs impose a different requirement. Fielded systems of the Sting class, developed by Wild Hornets against Shahed-type loitering munitions, are specified by their manufacturer at a maximum speed of 280 km/h, and test footage released in 2025 showed 315 km/h [5]; the targets they pursue cruise at 200–250 km/h. An interceptor must overtake what it chases, so its propulsor has to sustain axial flight an order of magnitude faster than multirotor cruise, for missions of a few minutes, on airframes that are expendable by design. A propeller optimised for low advance ratio then operates far from its design point, with the outboard sections approaching or entering negative incidence.

The expendable premise adds a manufacturing constraint. Injection moulding and carbon-fibre layup suit volume production but not rapid, distributed, low-volume manufacture. Additive manufacture suits that role, provided the blade is shaped for it. This note addresses both constraints on a parent blade whose geometry and measured performance are public.

2. Parent Propeller

The parent geometry is the DA4052 in its 9 × 6.75 in variant (229 mm diameter, p/D = 0.75), designed at UIUC using the PROFOIL inverse method with three 12 % thick sections along the blade [1,2,3]. It is adopted only as a starting point and as a future validation case: its geometry is published in tabulated form, its performance is independently measured, a three-blade configuration matching the modified rotor is reported, and it was itself 3D printed for the original campaign. Full parent data are not reproduced here; the subject of this note is the modified rotor.

3. Proposed Modification

Four geometric modifications are applied to the parent, together with a proposed shroud. They are coupled: the truncation is what permits the enlarged root chord, the thicker section and the higher twist to be carried without an unacceptable structural or inertial penalty.

M1 — Truncation. The blade is truncated at both ends. The tip trim removes the parent’s knife-edge outboard region, whose local thickness falls below 1 mm at this scale and which is the first part to fail in a printed blade; the root trim removes the narrow-chord dip. The 229 mm parent becomes a 170 mm rotor: a 60 mm blade on a 25 mm hub radius. Disc area, and therefore static thrust, is reduced, accepted as a deliberate trade in a design sized for high-speed axial flight rather than hover.

M2 — Root-chord override. The chord at the new root station is raised to 28 mm and ramped smoothly to the parent’s natural 41 mm peak, in place of the parent’s root dip. This places material where the bending moment is greatest rather than where the parent’s low-Reynolds optimisation had thinned it.

M3 — Section replacement. The 12 % PROFOIL sections are replaced throughout by the MH 113, a 14.6 % thick, 6.4 % cambered section from Hepperle’s MH 112–117 family of ultralight-aircraft propeller sections [4]. The family is graded in thickness for spanwise use, so a thinner member outboard remains open.

M4 — Twist increase. Twist follows constant ideal-pitch geometry, β = α + arctan(V/ωr), with α = 6° at the design condition, giving 69° at the root station and 36° at the tip (Figure 1). Geometric pitch at 75 % R is about 380 mm, so p/D = 2.23 against the parent’s 0.75. This modification defines the concept; M1–M3 make it realisable in a printed blade.

M5 — Cylindrical shroud (proposed). A plain ring around the rotor, with no lip contouring, diffuser or nozzle, is proposed for noise mitigation. It is deliberately the simplest possible shroud, so any acoustic benefit is attributable to the shroud rather than to a redesigned flow path.

The rotor has three blades and a solidity of about 0.24. It is analysed at 15,000 rpm, so that nD = 42.5 m/s. The design point is J = 1.8, that is 76.5 m/s or 275 km/h; the 300 km/h target corresponds to J = 1.96 and therefore sits slightly beyond it. Rotational tip speed is 133.5 m/s, and the helical tip Mach number is 0.44 at the design point.

Table 1. Summary of the modifications applied to the parent geometry.

Mod.	Change	Primary intent
M1	Tip + root truncation (229 → 170 mm)	Remove fragile tip; margin
M2	Root chord 28 mm, ramped to 41 mm	Material at max bending
M3	MH 113 section (14.6% t/c)	Printability, strength
M4	Twist 69°→36°; p/D 0.75 → 2.23	Design point at high J
M5	Cylindrical shroud	Noise (proposed)

Table 1. Summary of the modifications applied to the parent geometry.

Mod.	Change	Primary intent
M1	Tip + root truncation (229 → 170 mm)	Remove fragile tip; margin
M2	Root chord 28 mm, ramped to 41 mm	Material at max bending
M3	MH 113 section (14.6% t/c)	Printability, strength
M4	Twist 69°→36°; p/D 0.75 → 2.23	Design point at high J
M5	Cylindrical shroud	Noise (proposed)

4. Design Hypotheses

The following are the claims to be tested, not results.

H1. The modified blade attains higher peak propulsive efficiency than the parent at advance ratios corresponding to 83 m/s axial flight, at the cost of reduced static and low-J thrust.

H2. Reduced span combined with increased section thickness yields a positive structural margin at maximum operating rotational speed for an additively manufactured polymer blade over a mission of a few minutes.

H3. Under load the printed blade untwists: centrifugal and aerodynamic forces bend it towards a flatter pitch, so the blade that flies is not the blade that was drawn. The hypothesis is that this pitch loss is large enough that predictions made on the rigid CAD geometry must be corrected before they can fairly be compared with measurement.

H4. A plain cylindrical shroud reduces radiated tonal noise but produces a net propulsive-force penalty at the high-advance-ratio design point once shroud drag is accounted for.

H1 and H4 are addressed by computation and, ultimately, experiment; H2 and H3 by finite-element analysis with imported aerodynamic loading. H4 is stated in the form expected to hold; a neutral or beneficial net-force result would itself be reportable.

5. Preliminary Assessment

The modified rotor has been analysed in QBlade using blade-element-momentum theory at 15,000 rpm over advance ratios from 0.1 to 2.2. The predicted thrust is shown in Figure 2. It is nearly flat at about 41–44 N below 30 m/s, peaks at 43.7 N at 29.8 m/s, and then falls steadily. At the J = 1.8 design point the predicted thrust is 20.5 N and at 300 km/h it is 17.5 N; linear extrapolation of the computed trend places zero thrust near 116 m/s.

These are indicative figures only. Blade-element methods are least reliable in exactly the regime that defines this concept: high twist, low aspect ratio after trimming, and strong three-dimensional tip flow. The corresponding parent computation and the overlay against the published wind-tunnel data, which are what would establish the fidelity of the method on a known case, are not yet included; no measured performance is claimed.

6. Manufactured Article

The modified rotor has been manufactured as a single three-blade part by fused filament fabrication in PLA (Figure 3), with an integral deep-skirted hub. Print orientation, layer height, wall count and infill govern the anisotropy of the part and are reported as part of the specimen definition. The trim leaves a squared, unfaired tip, and the deposited build layers are visible on both faces in Figure 3; tip treatment is not part of the present concept and its aerodynamic and acoustic effect is left to the computational study. PLA is a prototyping material here, for geometry verification and low-speed bench work; the flight material remains open pending structural analysis. An ogival spinner of about 50 mm diameter is intended to fair the hub region and is not part of the article shown.

The rotor has been installed on a thrust stand and run on the bench (Figure 3c). The article is dimensionally sound, balances on the shaft and turns without contact, and a single static run produced 6 N of thrust. Rotational speed was not instrumented, and the throttle was not advanced further because rotor noise became the limiting factor at the test location. The measurement therefore establishes that the article runs and produces thrust, but it cannot be placed on the predicted curve of Figure 2, which requires a known rotational speed; no torque or acoustic measurement was taken. A controlled campaign with a tachometer, containment shielding and a stated uncertainty budget is required before any measured quantity is compared with prediction.

7. Limitations

No validated aerodynamic performance is claimed; the blade-element results are preliminary.

The modified geometry operates outside the advance-ratio range over which the parent was measured, so the reference data validate the method, not the operating point.

No structural analysis is reported; hypothesis H2 is stated but untested.

The single static measurement of 6 N was taken without a rotational-speed reading, so it cannot be compared with the prediction of Figure 2.

The shroud (M5) is a proposal; no aerodynamic or acoustic evaluation is presented.

Aeroelastic deformation is unquantified; predictions assume rigid geometry.

The blade-element results use incompressible section data; at a helical tip Mach number of 0.44 a 14.6 % thick, 6.4 % cambered section warrants a compressible check near the tip.

8. Ongoing Work

In progress for a subsequent full-length publication: RANS computation of the parent against the published data as a validation case, then a full advance-ratio sweep of the modified blade; one-way fluid-structure analysis across candidate print materials, including the deformed blade shape; a controlled static thrust campaign on the stand of Figure 3c; and evaluation of the shroud in net propulsive force and tonal noise.

9. Statement of Originality and Priority

The design concept described here — the truncation, root-chord override, section replacement, twist law and proposed cylindrical shroud applied to the DA4052 9 × 6.75 in parent geometry for high-advance-ratio interceptor propulsion — was conceived and developed by the author, who also produced the CAD definition, the manufactured article and the preliminary analysis reported here. This note is published to establish the date of the concept and its attribution. The DA4052 geometry and its wind-tunnel data are the work of the cited authors and are used as an open reference.