From 40 Minutes to 6 Hours: It's the Airframe, Not the Battery

🌏 中文版

The previous post computed multirotor hover endurance and closed with a debt:

No forward flight. Everything above is hover. Fixed-wing and VTOL aircraft generate lift from a wing in cruise, with a completely different power structure and endurance running to hours — that belongs to another post (the airframe-configuration cell).

This pays it. And the Taiwanese half turned out richer than I expected.

1. Two completely different power laws

A hovering multirotor generates all of its lift by pushing air:

P_hover = (m·g)^1.5 / sqrt(2·ρ·A) / (FM · η)

A fixed wing in level flight gets lift from the wing, so thrust only has to overcome drag — and drag equals weight divided by the lift-to-drag ratio:

P_cruise = (m·g / (L/D)) · v / (η_prop · η_elec)

The first goes as mass to the 1.5, the second as mass to the first power. And the second carries an extra divisor: L/D, which for small unmanned aircraft runs roughly 8 to 16.

Same 5 kg aircraft, same 2 kg pack (206 Wh/kg pack, 85% usable, 350 Wh):

Four times the power, four times the endurance, five times the range. Same weight, same battery.

This is not "fixed wings are better." The two are doing different things. What the multirotor buys with that 512 W is staying on this spot, which the fixed wing simply cannot do.

2. Fixed-wing range does not depend on how fast you fly

There is something easy to miss in that table: at L/D 8, whether you fly 15, 20 or 25 m/s, the range is 131 km.

Not a coincidence. Expand it:

range = speed × time = v × E / P
      = v × E / [ (m·g/(L/D)) · v / η ]
      = E · (L/D) · η / (m·g)

The speed v cancels. Fly faster and power rises proportionally while time falls proportionally; the two offset exactly.

This is the second result in this series, after the 2/3 battery optimum, where every parameter cancels out. An electric fixed wing's range depends on exactly four things: how much energy it carries, how good its L/D is, how efficient its propulsion is, and how heavy it is. How fast you fly does not affect how far you can go.

(The honest caveat: real L/D varies with speed and peaks somewhere, falling off if you fly too slow or too fast. So in practice there is still a best cruise speed. But within the range where L/D is roughly flat, "flying faster saves time, not energy" holds.)

3. A VTOL's cost is not the transition, it is carrying two systems

A VTOL wants both: vertical takeoff with no runway, and fixed-wing cruise efficiency. Where does it pay?

Intuition says the energy burned in transition. Compute it: the 5 kg aircraft above hovers on 512 W, and if takeoff and landing transitions total 60 seconds, that is 8.5 Wh — 2.4% of a 350 Wh pack. Negligible.

The real cost is mass. A VTOL must carry both a rotor system (motors, ESCs, propellers) and a wing, so its structure is heavier than a pure fixed wing's. Same 5 kg takeoff weight, same 1.5 kg payload, with the VTOL airframe 0.5 kg heavier:

That extra 0.5 kg of structure comes straight out of the battery, and range drops 25%.

So the right way to understand a VTOL is not "both worlds" but trading roughly a quarter of your range for not needing a runway. Whether that is a good deal depends entirely on whether you have a runway — in Taiwan's mountains, outlying islands, and on ships, that quarter usually buys very well.

4. Why a single-rotor helicopter beats a multirotor: disk area

In hover, a helicopter's endurance is clearly better than a multirotor's. The reason is the denominator of the first equation: sqrt(2ρA). A is disk area, and bigger is cheaper.

A 25 kg aircraft carrying the same 8 kg battery:

Same weight, same battery — swapping four small propellers for one large rotor doubles the endurance.

The price is mechanical complexity: a collective-pitch main rotor, a tail rotor to counter torque, and a drivetrain full of parts that wear. The two TTSB occurrences taken apart in the crash anatomy post failed exactly there — once in the main rotor servo's electrical system, once in a fractured tail-rotor pitch link. A multirotor has none of those parts because it has none of those mechanisms.

Endurance and mechanical simplicity are a trade, and those two accident reports are the bill for it.

5. Taiwan's endurance ladder: from AXH-E230RS to T-400, a domestic aircraft at every rung

Before writing this I assumed Taiwanese drones were nearly all multirotors. They are not — there is a domestic aircraft at every rung, and the way the ladder climbs maps exactly onto the two variables computed above: disk area and energy source.

(The "over 25 kg" for the AXH-E230RS is an inference, not a published figure — it appears in the TTSB accident statistics, and that statistic's threshold is "maximum takeoff weight exceeding 25 kg." The remaining electric figures are manufacturer-published; the fuel-cell and T-400 figures come from a Ministry of Economic Affairs Department of Industrial Technology exhibition report, sourced to ITRI's ITIS research team in September 2023, so a snapshot of that date.)

Three things worth separating out.

First, DRAGONFLY validates the previous section's model. I computed 148 km for a 5 kg VTOL; this 13 kg aircraft is rated at 150 km. The orders of magnitude line up, and its method is exactly tiltrotor — propellers pointing up for takeoff and hover, rotating forward for cruise once at altitude.

Second, fuel cells buy a factor of three. The previous post said hydrogen fuel cells were out of scope; here is the answer. In the same 25 kg class, an electric single-rotor helicopter does about 60 minutes; on a fuel cell it does 3 hours. ITRI also flew real long-range validations — Magong to Wang'an and back, 34 km with a 6.6 kg payload; supplies to the Xinda mountain hut at 3,200 m; and Tainan to Dongji Island and back, 88 km across open water.

Third, the piston rung is different physics. Thunder Tiger's T-400 uses an Austrian Rotax 912 aero engine, a 4.4 m main rotor, and flies 6 hours. Gasoline's energy density is roughly forty times a lithium cell's, and even at 30% engine efficiency what remains still dwarfs the battery. To reach the "hours" scale, what you change is the energy source, not the battery. (Thunder Tiger is the company whose public filings an earlier post took apart.)

6. That 24.9 kg

The ITRI and AVIX Tech fuel-cell light helicopter lists a maximum takeoff weight of 24.9 kg.

The spec-sheet post noted manufacturers pinning weights at 249 g and 1.99 kg, because 250 g and 2 kg are each a legal threshold. 24.9 kg is the same behaviour pressed up against the 25 kg line — above 25 kg requires physical inspection, and is also what enters the TTSB accident statistics.

This is not a problem in itself; if anything it shows the regime working. Regulatory thresholds really do reach back and shape a machine's design parameters, right down to a national research institute's development platform. A difference of 0.1 kg puts you under an entirely different inspection procedure.

7. So why are Taiwan's applications almost all multirotor?

Physics gives fixed wings a five-fold coverage advantage, yet the Taiwanese applications this series has taken apart — agricultural spraying, inspection, search and rescue, logistics — use rotorcraft almost exclusively. Two reasons, neither of them preference.

First, mission shape determines configuration. A fixed wing's advantage is covering large areas, and the shape of these missions is "go to a point and stay there": bridges want stationary photography, transmission towers want orbiting, fields want slow low passes, mountain rescue wants threading through trees, island delivery is a few-kilometre hop. Not one of them is "sweep 200 km² of open country" — which is where fixed wings actually win, and which Taiwan's terrain and application structure naturally produce very little of.

Second, regulation eats the fixed wing's advantage. A fixed wing exists for that 197 km of range, but the BVLOS post concluded Taiwan has no beyond-visual-line-of-sight framework, so genuinely long flights only happen through case-by-case approval. You cannot fly the 197 km, so the 197 km is not worth anything. Whereas a multirotor's "stay on this spot" works perfectly within visual line of sight — which is exactly what the inspection post meant by routing around BVLOS, because that work segments naturally.

So "Taiwan uses a lot of multirotors" is not evidence of falling behind, it is the reasonable outcome of mission structure plus regulatory structure. And the fuel-cell and piston rungs appearing mainly on research platforms and defence models follows the same logic: the buyers for that capability are the public sector and defence, not commercial applications.

What this post does not answer

• No climb or wind performance. Everything above is level cruise or stationary hover. Climbing to 3,000 m, or holding position in a force 6 wind, has a different power structure — and Taiwan's mountain and coastal missions involve both.
• No test conditions for the aircraft cited. The endurance and range figures above are manufacturer or research-institute published values, and the previous post just demonstrated how much the fine print under "max flight time" can matter. Without test conditions these cannot be compared to each other, and I do not have their fine print.
• No costs. The endurance fuel cells and piston engines buy is real, but unit price, maintenance, and the logistics of hydrogen or fuel are entirely uncosted. Three hours of endurance paired with a site that cannot source hydrogen is three hours on paper.
• No tethered aircraft. Run a power cable up and it flies for a very long time. That is a fifth answer, at the price of not being able to move at all.

References

Computation

Hover power P = (m·g)^1.5 / sqrt(2ρA) / (FM·η), cruise power P = (m·g/(L/D))·v / (η_prop·η_elec), ρ = 1.225 kg/m³. Parameters: FM 0.70, motor and ESC efficiency 0.80–0.85, propeller efficiency 0.75, L/D at 8 / 12 / 16; pack 206 Wh/kg, 85% usable. Range's independence from speed follows directly from R = E·(L/D)·η/(m·g).

Primary: aircraft specifications

• AXH-E230RS unmanned helicopter — Wikipedia (AVIX Tech; single main rotor plus tail rotor, electric brushless motors; the article cites manufacturer-published data: about 30 km maximum control distance, about 85 km/h cruise, about 4.6 kg maximum external payload, up to about 60 minutes endurance, operable up to Beaufort force 6) (in Mandarin)
• Ministry of Economic Affairs, Department of Industrial Technology — technology briefing on unmanned aerial vehicles (ITRI fuel-cell twin helicopter at 63 kg / 10 kg payload / 2 hours; the ITRI and AVIX Tech fuel-cell light helicopter at 24.9 kg / 5 kg payload / 3 hours; validation flights Magong–Wang'an 34 km round trip, Xinda hut at 3,200 m, Tainan–Dongji 88 km round trip; Zang Shi Technology's DRAGONFLY twin tiltrotor at 1.5 m span / 13 kg / 150 km / 1.8 hours; Thunder Tiger's T-400 with a 4.4 m main rotor / Rotax 912 / 180 kg / 250 km control distance / 6 hours. Sourced to ITRI's ITIS research team, September 2023) (in Mandarin)

On this site

• Why Drones Only Fly 30 to 45 Minutes: One Equation Gives the Answer
• Taking Apart Two TTSB Crash Reports: Neither Was the Operator's Fault
• BVLOS in Three Jurisdictions: Taiwan Has No Framework At All
• Inspection Is Taiwan's Furthest-Along Drone Application — Because It Routed Around BVLOS
• How to Read a Drone Spec Sheet: Which Lines Regulation Turned Into Boundaries
• The Filings Answered What I Assumed Needed an Interview