Case Study: CFRP Actuator Housing for eVTOL Tilt-Rotor — 41% Mass Reduction with DO-160G Vibration Qualification

When a European eVTOL OEM approached us to re-engineer the tilt-rotor actuator housing for their 2.5-tonne MTOW aircraft, the target was aggressive: cut mass by at least 35% versus the incumbent 7075-T6 aluminum design while passing RTCA DO-160G Section 8 vibration and maintaining ±0.05 mm interface tolerances. This CFRP actuator housing case study details how a hybrid carbon fiber/epoxy architecture achieved a 41% mass reduction — from 1.84 kg to 1.09 kg — and survived 12 hours of random vibration at 14.1 Grms without delamination or fastener pull-through. We cover material selection, laminate design, FEA validation, CNC machining strategy, and the qualification test matrix that closed the program in 19 weeks.

Baseline Aluminum Housing: Performance Envelope and Mass Penalty

The incumbent housing was machined from 7075-T6 aluminum billet (UTS 572 MPa, yield 503 MPa, density 2.81 g/cm³). It housed a planetary gearset and a brushless DC motor driving the tilt-rotor nacelle through a 4.2:1 reduction. The part carried three load cases:

• Limit torque reaction: 312 N·m at the actuator output flange
• Vibration: DO-160G Category R, curve B (14.1 Grms, 10–2000 Hz)
• Thermal: -55°C to +85°C operating, +105°C ground soak

Mass was 1.84 kg after finish machining, with wall thicknesses of 3.0–4.5 mm. The OEM's mass budget for the redesigned housing was 1.20 kg — a 35% reduction — with no compromise on stiffness at the output bearing bore (max deflection 0.08 mm under limit load).

Why CFRP Actuator Housing Architecture Beats Aluminum for eVTOL

Aluminum's isotropic stiffness (E = 71 GPa) forces designers to add material where bending moments peak. Carbon fiber reinforced polymer (CFRP) allows directional reinforcement — fibers aligned to principal stress vectors — so material is placed only where it resists load. For this housing, the dominant stress path runs from the output bearing bore to the three airframe attachment lugs. A quasi-isotropic laminate would waste mass; a tailored layup recovers 2.5–3.5× the specific stiffness of 7075-T6.

We selected Toray T700S carbon fiber (4,900 MPa tensile, 230 GPa modulus) with Hexcel 8552 epoxy (Tg > 190°C, Vf > 62%). T800H (5,490 MPa, 294 GPa) was evaluated but rejected on cost — the stiffness gain did not justify a 38% fiber price premium for this load case.

The 11× lower CTE in the fiber direction also reduced thermal distortion at the bearing bore — a secondary win for the ±0.05 mm tolerance stack.

Laminate Design and FEA Validation

The housing was modeled as a 16-ply laminate with a tailored stacking sequence: [0/±45/90/0₂/±45/0/90]ₛ. Ply thickness was 0.185 mm (T700S/8552 unidirectional tape), giving a nominal wall of 2.96 mm. Local pad-ups to 24 plies (4.44 mm) reinforced the attachment lugs and bearing bore.

We used ANSYS Composite PrepPost with a Tsai-Wu failure criterion. Under limit load (312 N·m), the maximum inverse reserve factor (IRF) was 0.62 — a safety margin of 1.61. Under ultimate load (1.5× limit), IRF peaked at 0.89, still below 1.0. The first natural frequency shifted from 412 Hz (aluminum) to 387 Hz (CFRP), a 6% drop that remained well above the 200 Hz minimum required to avoid coupling with rotor harmonics.

Worked mass calculation: The aluminum housing displaced 655 cm³ of material (1.84 kg ÷ 2.81 g/cm³). The CFRP design displaced 690 cm³ (1.09 kg ÷ 1.58 g/cm³) — slightly more volume, but 41% less mass because of the density difference. The mass saving comes entirely from material density and directional stiffness, not from shrinking the geometry.

CNC Machining and Autoclave Process

We laid up the laminate on a 6061 aluminum mandrel, vacuum-bagged, and cured in an autoclave at 135°C and 6 bar for 180 minutes per Hexcel 8552's recommended cycle. Post-cure, the blank was CNC-machined on a DMG Mori 5-axis mill to final dimensions. Critical features — bearing bore, lug holes, and sealing faces — were machined to ±0.05 mm and verified on a Zeiss Contura CMM. Surface finish on the bore was Ra 0.8 µm.

Fastener holes were drilled with diamond-coated tooling at 0.05 mm/rev feed and 80 m/min cutting speed to avoid delamination. We used a 4-mm peck drill cycle and backed the laminate with a sacrificial aluminum plate. Hole quality was verified per ASTM D5961 (bearing strength) on coupons; average bearing strength was 412 MPa, exceeding the 380 MPa design allowable.

DO-160G Vibration Qualification Results

The housing was instrumented with 12 strain gauges and three accelerometers and mounted on a shaker table. The test profile was DO-160G Section 8, Category R, Curve B: 14.1 Grms random vibration from 10 Hz to 2000 Hz, 12 hours per axis (36 hours total).

Results:

• No delamination detected by ultrasonic C-scan after 36 hours
• Maximum strain recorded: 1,840 µε at the lug root — 34% of the 5,400 µε design allowable
• First natural frequency shifted by less than 2% post-test (387 Hz to 380 Hz)
• No fastener pull-through or bearing bore elongation > 0.01 mm

The housing passed DO-160G with margin. The aluminum baseline had previously passed the same profile but with a 1.84 kg mass penalty.

Program Outcome and Production Status

The CFRP actuator housing entered production in Q1 2025. Key metrics:

The 41% mass reduction contributed to a 2.3% overall aircraft mass saving, translating to an estimated 4.1% range improvement on the OEM's 120 km mission profile. The program ran from kickoff to qualification in 19 weeks.