Icing Certification Testing for Large Fixed-Wing UAS

Icing on a small electric UAV is usually solved with a heated propeller or a heated pitot tube — a targeted fix for a small, power-constrained component. Icing on a MALE or HALE aircraft is a different order of problem entirely: it's a full-aircraft, standards-driven certification requirement, tested against the same icing envelope manned transport aircraft are certified against, because at this scale the vehicle is flying the same weather.

1. The certification standard is the same one manned aircraft use

The reference facility for this class of testing is NASA's Icing Research Tunnel (IRT) at Glenn Research Center — one of the world's largest refrigerated wind tunnels dedicated to aircraft icing, and a facility that explicitly replicates the FAA icing certification standard in Federal Aviation Regulation Part 25, Appendix C . Its published characteristics show what "full-scale icing certification" actually requires:

Test section 6 ft high × 9 ft wide × 20 ft long
Test section air velocity 50 to 350 knots
Test section total temperature –25°C to 5°C
Liquid water content 0.2 to 3.0 g/m³ , with cloud uniformity held to ±10%
Droplet median volume diameter 15 to 50 µm
A uniform icing cloud covering 4.5 by 6 feet of the test section
Simulated engine bleed-air flow up to 1 lb/s , for testing thermal anti-icing systems that use hot bleed air

The IRT tests full-size aircraft components as well as scale models of both airplanes and helicopters, and its customer list — Boeing, Bell Helicopter, Goodrich, Cessna, Sikorsky, plus the U.S. Army, Navy and Air Force — reflects that this is certification-grade infrastructure, not a research curiosity. More recent NASA icing test campaigns have also extended calibration work toward the newer Appendix O supercooled large drop (SLD) conditions that both FAA and EASA now require alongside the original Appendix C envelope.

2. Why the ice-protection technology itself diverges by scale

Ice protection technology splits cleanly along aircraft scale, and it splits for a physical reason: available power. Electrothermal heating — the approach typical of small UAVs — is limited by the battery amperage a small electric platform can supply, which is why it's usually confined to small, high-value components like a propeller or a pitot probe rather than an entire wing.

A large, fuel-powered MALE or HALE aircraft has a different power budget entirely, and that opens up thermal anti-icing using engine bleed air , ducted to the wing leading edge through a piccolo tube — the same basic architecture manned transport aircraft have used for decades, and one that saves the aircraft from carrying the electrical generation capacity a fully electric ice-protection system would demand. Documented examples at this scale include the RQ-4 Global Hawk , whose engine cowl anti-icing system draws heat extracted from the aircraft's hydraulic system rather than bleed air — a reminder that "thermal protection" on a large UAS can be engineered from more than one onboard heat source, and the right choice depends on the aircraft's specific systems architecture.

3. What a large-fixed-wing icing test campaign needs to cover

Testing against the actual certification envelope , not an arbitrary cold-and-wet condition — Appendix C liquid water content, droplet size and temperature combinations are specified precisely because different combinations produce different ice shapes and different aerodynamic penalties.
Appendix O supercooled large drop conditions alongside Appendix C , since SLD icing produces ice accretion further aft on the airfoil than classical Appendix C icing, in a location ice-protection systems designed only against Appendix C may not cover.
The whole thermal system, not just the heated surface — for a bleed-air system, that means testing the ducting, piccolo tube distribution and heat delivery under simulated engine flow, not just measuring ice shed from a heated panel in isolation.
Aerodynamic performance with realistic ice shapes attached , since certification is ultimately about the aircraft's flying qualities with the ice the protection system is designed around — not just whether the protection system prevents ice formation in the best case.
Full-size components wherever the certification basis calls for them — the IRT's documented capacity to test full-size aircraft components exists precisely because ice accretion physics on some components does not scale cleanly to a small model.

4. Five mistakes we see

Treating icing protection as a component-level electrothermal problem at any aircraft scale. What works for a small UAV's propeller does not extrapolate to a MALE/HALE wing — the power budget and the certification basis are both different.
Testing only Appendix C and assuming that covers "icing." Appendix O supercooled large drop conditions are now a distinct, required part of the certification basis for many programmes, and they ice a different part of the airfoil.
Validating the heat source without validating the delivery system. A bleed-air system's piccolo tube distribution and ducting losses matter as much as the raw heat available — a system that works on paper can under-deliver at the wingtip.
Measuring ice prevention without measuring aerodynamic penalty. Certification isn't just "does ice form" — it's "how does the aircraft fly with the ice shapes the protection system is validated against."
Assuming a scaled model captures everything a full-size component would. Ice accretion physics is one of the areas where scale effects are least forgiving, which is exactly why certification-grade icing tunnels test full-size components directly.

5. What this means for your programme

Tell us your certification basis and ice-protection architecture , and we'll scope a test programme against the actual Appendix C and Appendix O envelope your aircraft needs to certify against — not a generic cold-weather check. Related reading: our hub article on

long-range fixed-wing UAS testing

Sources and notes

NASA Icing Research Tunnel facility characteristics (test section 6 ft × 9 ft × 20 ft, air velocity 50–350 knots, total temperature –25°C to 5°C, liquid water content 0.2–3.0 g/m³ with ±10% cloud uniformity, droplet MVD 15–50 µm, uniform icing cloud 4.5 by 6 ft, simulated engine bleed-air flow up to 1 lb/s, replication of FAA FAR Part 25 Appendix C icing certification standards, testing of full-size aircraft components and scale models of airplanes and helicopters, and its customer base including Boeing, Bell Helicopter, Goodrich, Cessna, Sikorsky and U.S. military services): NASA Glenn Research Center / NASA Aeronautics Test Program, "Icing Research Tunnel" fact sheet.
Extension of IRT calibration work to FAA/EASA Appendix O supercooled large drop (SLD) conditions in more recent NASA test campaigns is documented in subsequent NASA Glenn icing research publications on the NASA Technical Reports Server.
Electrothermal ice protection as the typical approach for small, power-constrained UAV components, versus thermal bleed-air anti-icing via piccolo tube as the approach suited to large, fuel-powered HALE aircraft, and the RQ-4 Global Hawk's engine cowl anti-icing system drawing heat from the aircraft's hydraulic system: consistent with published aircraft ice-protection-system literature.
Aircraft background for RQ-4B Global Hawk (HALE class): consistent with our published article on long-range fixed-wing UAS testing (MALE, HALE and VTOL).
Test-methodology guidance (full-envelope Appendix C and Appendix O testing, whole-system thermal validation, aerodynamic-penalty measurement with ice shapes attached, full-size component testing) is TunnelTech engineering practice, directly responsive to the certification-grade facility characteristics documented above.

Icing and All-Weather Certification for Large Fixed-Wing UAS