KDE 5215XF-220 or MAD V8013 PRO for the pusher?

The short answer

Fly the MAD, and plan on running it from the 14S pack.

It is the only one of the two rated for PT2's 14S Longshot pusher pack. It uses less battery for the same thrust, and it spreads its heat through twice the metal. The KDE is not a bad pusher on a 12S pack: it is lighter and keeps more top-end speed as the pack drains. It just has little thermal margin in a spot with poor airflow, and no path to 14S.

What the pusher has to do

At 25 m/s the wing needs a lift coefficient of to carry 25 kg. The drag-corrected ADB polar puts drag there at N, so the pusher has to deliver W of thrust power. The 20×12 runs at an advance ratio of and about % efficiency, so the motor turns it at rpm with N·m and W at the shaft. That shaft work is the same for both motors. What differs is what each one burns to make it.

Two very different motors

The KDE is small and fast: 220 Kv, 0.043 N·m per amp, 305 g. The MAD is large and slow: 150 Kv, 0.064 N·m per amp, 591 g. To make the same N·m, the KDE needs A and the MAD A. Copper loss grows with the square of current, and the KDE also has twice the winding resistance, so it makes several times the MAD's copper loss. The MAD gives some of that back as iron loss: its big stator burns about W just spinning at rpm, well above the 3,000 to 4,500 rpm it was designed to hover at.

Can it reach the rpm? The pack decides

A 12-inch pitch at 25 m/s needs about rpm. On a full 12S pack both motors get there with room to spare, and their full-throttle thrust at 25 m/s is about even. As the pack drains, the MAD, at 150 Kv, loses top end faster than the KDE. With the expected N it still holds % margin at 20% charge. With the 30 N design case that drops to %, against % for the KDE. On 14S the MAD has plenty of room. The KDE can't go there: KDE rates the 5215XF to 12S LiHV (52.2 V) and its UAS55 HVC ESC to 55 V, and a full 14S pack is 58.8 V.

Thrust across the speed range

Full-throttle thrust from each motor against the drag the aircraft needs at each speed. Where a motor's curve crosses the drag curve is its top level speed on this pack. The scenario bar above changes the pack, charge, prop, drag case and air; every figure from here down follows it.

Heat, and why the pusher spot is hard

A lift motor sits in its own prop's downwash. A pusher motor sits at the hub of a prop that pulls air toward it, right behind a fuselage that blocks the free stream. The hub is where the prop moves the least air. KDE rates the 5215XF at 44 A for 180 s, and only with at least 5 mph of airflow across the windings; there is no continuous rating. MAD rates the V8013 as sustainable below 24 A with prop-wash cooling at 25 °C, and 24 to 78 A for 10 to 30 s.

Which edge of each band is right, and how much air the motor really gets, can't be settled from datasheets. KDE's 180 s rating only works out if its motor sheds heat 2 to 3 times better per square centimetre than MAD's sustainable rating assumes for the V8013. That fits KDE's open, through-flow design, but only when air actually passes through the windings, which is exactly what a pusher at the end of a tail cone may not get. Two things hold either way. The KDE makes its heat in half the mass, so it heats 3 to 4 times faster whenever the airflow isn't there: transitions, ground runs, a buried mount. And KDE publishes no continuous rating at all. Measure it on the first ground runs (step 4 in section 10).

Transition and ground runs

The hardest minutes for the pusher are the transition and the bench. At zero airspeed the 20×12 is a heavy load, and in a pusher the motor gets the least cooling there.

Battery use across the speed range

Pack power in level flight at each speed, and what it means for range. The difference between the motors is small next to the choice of cruise speed: this airframe's best lift-to-drag sits near m/s.

Each motor wants a different prop

The KDE wants less torque and more rpm, so a smaller or flatter prop. The MAD wants more torque and fewer rpm, so a bigger or steeper one, which also cuts its iron loss. The 20×12 sits between them. APC data stands in for the wooden props, so check real wood props against the ground test before trusting small differences.

Before PT1.5 flies

Assumptions, sources and open questions

Sources