Sizing a multirotor from scratch

The loop you are actually solving

Sizing a multirotor is circular, and pretending otherwise is why first attempts go wrong. The battery you pick changes the all-up weight, the all-up weight changes the thrust you need, the thrust changes the motor, and the motor changes the battery. There is no starting point that avoids this. What there is is a sensible first guess and two or three iterations that converge quickly.

Work in this order, then go round again:

1. All-up weight, guessed
2. Thrust required
3. Motor and propeller together
4. ESC, from the motor current
5. Battery, from the flight time you want
6. Back to step 1 with the real battery weight

Step 1: all-up weight

All-up weight (AUW) is everything that leaves the ground: frame, motors, propellers, ESCs, flight controller, battery, payload, cabling, screws. The last three are the ones people forget, and together they are rarely less than 10% of the total.

If you are copying a known design, start from its published AUW. If you are not, add up what you know and add 15% for the things you have not thought of. It will still be light, but it will be close enough to choose a motor from.

Step 2: thrust required

The number that matters is the ratio of total available static thrust to all-up weight, at full throttle:

thrust-to-weight = (thrust per motor × number of motors) ÷ AUW

A ratio of 1 means the aircraft can just barely hold itself in the air at full throttle, with nothing left to manoeuvre, correct for wind, or recover from a mistake. Nobody flies that.

2:1 is the working minimum for a stable camera platform or a heavy lifter. It means hovering happens at about half throttle, which matters for a reason that is not obvious: a motor near half throttle is close to its efficient region, and a motor at 85% throttle is not. Flying a badly sized aircraft costs endurance as well as authority.

Freestyle and racing quads run 6:1 to 12:1, because there the point is acceleration rather than efficiency. Heavy-lift and survey aircraft sit near 2:1 and stay there.

Step 3: motor and propeller, together

These are one decision, not two. A motor produces thrust only through a propeller, and every thrust figure worth reading names the propeller it was measured with. A motor quoted at "2.4 kg thrust" with no propeller, no cell count and no throttle setting has told you nothing.

Divide the thrust you need by the number of motors, then look for a motor that reaches it with margin, on the cell count you intend to run.

Stator size sets what the motor can physically swing. The code is diameter and height in millimetres: a 2207 stator is 22 mm across and 7 mm tall. Bigger stators move more air and take bigger propellers. KV is a separate axis: revolutions per volt with no load. High KV spins a small propeller fast, low KV turns a large one slowly, and the large slow one is almost always more efficient.

Step 4: the ESC

The speed controller has one job that can kill it: passing the motor's current without overheating. Size it from the motor's maximum current draw, not its typical one, and add headroom.

ESC continuous rating ≥ motor max current × 1.2

Check the cell count too. An ESC rated to 6S will not survive 8S, and the failure is immediate rather than gradual. If a motor is going to see 65 A on a 6S pack, a 60 A ESC is not marginal, it is wrong.

Step 5: the battery

Three things to get right, in this order.

Voltage comes from the cell count, and it has to match what the motor and ESC expect. A 6S LiPo is 22.2 V nominal and 25.2 V charged.

Current: the pack has to supply everything the motors draw at once. Multiply the motor maximum by the number of motors, then check the pack can deliver it:

pack max current = capacity in Ah × C rating

Energy sets endurance, and energy is watt-hours, not milliamp-hours. See why mAh is the wrong number.

Step 6: go round again

You now know the real battery weight. Put it back into the all-up weight and redo step 2. Usually the answer barely moves and you are finished. Sometimes the battery is heavy enough to push the thrust ratio under 2:1, and then you have a genuine decision to make: a bigger motor, a smaller pack, or a lighter frame. That decision is the whole design, and it is why the loop exists.

A worked example

A 5-inch freestyle quad, built around parts that exist:

• Guessed AUW: 650 g with a 6S 1300 mAh pack
• Target ratio: 8:1, so 5.2 kg total, or 1.3 kg per motor
• Motor: a 2207 in the 1700 to 1900 KV range on 6S reaches about 1.5 kg on a 5 inch tri-blade. Margin, so it stands.
• Propeller: 5.1 x 4.4 tri-blade, which is what that stator is built around
• Motor max current: roughly 35 A, so a 45 A or 50 A 4-in-1 ESC rated to 6S
• Battery: 4 × 35 A is 140 A peak. A 1300 mAh pack needs 1.3 Ah × C ≥ 140 A, so about 110C. A 6S 1300 mAh 120C pack is a real product and weighs around 200 g, which is what was guessed.

The loop closed on the first pass, because the initial guess came from a known design. That is the advantage of starting from something that flies.