Design wings from measured data, make a glider stable and controllable in all three axes, and prove it with models, a wind tunnel and unmanned tests before anyone rides it.
A replica of the Wright 1902 glider in a museum: biplane wings, the elevator in front, the rudder behind, the pilot lying on the lower wing.Wright Brothers 1902 glider · 2009-01-02 · Jarek Tuszyński · CC BY 3.0 · Commons ↗
Otto Lilienthal made more than 2,000 glides, steering by swinging his body, and died in 1896 after one stalled in a gust (Britannica). The Wright brothers' 1901 glider gave only a third of the lift the published tables predicted (NASA), so they built a wind tunnel, measured more than 200 wing surfaces in about two months, and designed their 1902 glider from their own data. It made nearly 1,000 glides and solved control in all three axes (National Park Service). The order they worked in is the order on this page, because each stage was paid for in crashes.
The shape of it
Measure before building: lift and drag of each wing shape, in a wind tunnel.
Long, thin, gently curved wings: span about six times the chord, curved about 1 in 20.
Stable in pitch: balance ahead of the neutral point, and a tail or elevator to hold it there.
Control in three axes: pitch, roll and yaw, each tested.
Test without a person first: models, then the full glider flown as a kite.
People fly only proven designs, after training.
Wright 1902 glider: general arrangementFig 1
TimberLoose fill / grain
Span 9.78 m, chord about 1.5 m, length 4.90 m; elevator in front for pitch, a hip cradle that warps the wings for roll, and a movable rudder behind for yaw. From the National Park Service's figures; not a construction drawing.
Doing it, step by step
The summary above says what has to happen. This says how, with the numbers that decide whether it works.
1 — Measure before you build
The Wrights built their 1900 and 1901 gliders from Lilienthal's published tables; the 1901 glider gave only a third of the lift predicted (NASA). So they built a wind tunnel: a wooden box 6 ft (1.83 m) long and 16 in (41 cm) square inside, with air at about 25 mph (40 km/h) (National Park Service).
In about two months at the end of 1901 they tested more than 200 surfaces, at angles from 2° to 45°: aspect ratio, camber, the shapes of leading and trailing edges and wing tips (National Park Service).
Their model wings were strips of 20-gauge steel, cut, hammered, filed and soldered (NASA).
✓ Check: The lift and drag of your wing section are measured across the angles you will fly at, and recorded.
2 — The wing: long, thin and gently curved
Aspect ratio: the tunnel showed the value of a long, narrow wing, and the Wrights made the 1902 glider's span about six times its chord, against three in 1901 (National Park Service). A long wing loses less of its lift to the swirl at its tips.
Camber: a gentle curve front to back; the 1903 wing curved 1 in 20.
The glide: the 1902 glider's best glide was at 5°, about 11 m forward for every 1 m of height, twice as efficient as any glider before it (National Park Service).
✓ Check: The wing's span is at least six times its chord, and its measured lift at flying speed carries the whole weight with a margin.
3 — Stable in pitch: balance ahead of the neutral point
An aircraft turns about its centre of gravity. It is stable in pitch only if its neutral point — where the lift of wing and tail together effectively acts — lies behind the centre of gravity (NASA). Then a gust that raises the nose adds lift behind the balance point and pushes the nose back down.
A tail or elevator makes it so; the further back and the bigger, the more stable.
Balance too far aft and the aircraft pitches up, stalls and falls; too far forward and it dives and cannot raise its nose to land.
Find the limits with models: the model glider shows both in a minute.
✓ Check: A model with the same balance glides steadily and returns to its glide after a push.
4 — Control in three axes
Pitch (nose up and down): the elevator. The Wrights put theirs in front and made it movable.
Roll (one wing up): the Wrights warped the wing tips, twisting one up and the other down; on the 1902 glider the pilot did it by sliding his hips in a cradle (National Park Service). Ailerons do the same job today.
Yaw (nose left and right): a movable rudder behind. Warping to roll also swings the nose the wrong way; on the 1902 glider a movable rudder cured it, and on the 1903 machine the rudders were linked to the warping (National Park Service).
Lilienthal and the gliders before him steered by shifting the pilot's body, which cannot recover a stall in a gust.
✓ Check: Each control has been worked on the ground and on the tethered glider, and moves it the way intended.
5 — Fly it as a kite before anyone rides
1899: a biplane kite of 5 ft (1.5 m) span, its wings warped by four cords from the ground, proved the Wrights' roll control before any glider existed.
1900: they flew their first full-size glider as a kite, held by two ropes, its balancing worked by cords from the ground, with chain as ballast in light winds; only then did they make about a dozen free glides down the side of a 100 ft sand dune (National Park Service).
Low, into a steady wind, over soft sand: the first glides are short and close to the ground.
✓ Check: The full-size glider has flown empty on lines and answered each control from the ground before anyone rode it.
6 — Gliders today, and who flies them
Modern sailplanes are built of composites, with long, thin wings, and climb in rising air; hang gliders and paragliders are flown from hills. Every one is flown by a pilot trained at a school, in a proven, certified or inspected aircraft (FAA Glider Flying Handbook).
This site draws the 1902 glider as a general arrangement and gives no construction procedure for any glider that carries a person. The recognised route is a proven design built from its designer's plans or kit, inspected as it is built (FAA AC 20-27G), and flown only after training.
A modern sailplane on the grass at Borkenberge airfield, Germany, 2014: long, thin wings for a flat glide.Lüdinghausen, Flugplatz Borkenberge -- 2014 -- 2466 · Taken on 12 April 2014, 10:14 · Dietmar Rabich · CC BY-SA 4.0 · Commons ↗
✓ Check: Anyone who will fly has trained at a school, on a proven glider.
What it looks like
A hand-launched model glider a moment after the throw, climbing out over water.Handlaunched Glider · Summer 2005date QS:P,+2005-00- · Tord55 · Public domain · Commons ↗A modern sailplane on the grass at Borkenberge airfield, Germany, 2014: long, thin wings for a flat glide.Lüdinghausen, Flugplatz Borkenberge -- 2014 -- 2466 · Taken on 12 April 2014, 10:14 · Dietmar Rabich · CC BY-SA 4.0 · Commons ↗
Reference sheets
One drawn sheet covering the whole subject at once — the breadth a photograph cannot give, bought at the cost of accuracy. Read them for range and vocabulary, never as the thing you identify from.
Wind tunnel: measuring lift and dragThe parts of a small open wind tunnel, how lift and drag arise, a force balance on load cells, common test models, what is measured, lift and drag against angle of attack, smoke to show the flow, and a six-step build.Illustration, not a photograph.The fan is drawn at the inlet, blowing straight into the test section, with nothing to straighten the flow. Its swirl and gusts then reach the model, and every reading is wrong. Put the fan downstream, drawing air in through a rounded bell mouth, a honeycomb and a fine screen, as the wind tunnel does. The diffuser does not smooth the flow: it slows the air gently after the test section so the fan needs less power. Burn incense or make smoke only in a ventilated room.Wind tunnel: measuring lift and drag · 2026-09-29 · Civilization101 Blueprint (AI-generated) · CC BY-SA 4.0
Build it
Dimensioned drawings, bills of materials and construction procedures for the things this step asks for.
What the record shows was done at this step, what to do instead at this tier, and the evidence. All of them: the better path.
Adopt earlier
Measure in a wind tunnel before building
What was done
Early experimenters built full-size machines from published tables and learned from crashes; the Wrights built their tunnel only after their 1901 glider fell far short.
Do instead
Build a small wind tunnel and measure every wing shape before cutting a spar.
Evidence
The 1901 glider gave a third of the lift predicted (NASA); after two months in the tunnel, the 1902 glider made nearly 1,000 glides (National Park Service).
Designing from someone else's tables without measuring: the Wrights' 1901 glider gave a third of the lift predicted.
Balance too far aft: the glider pitches up, stalls and falls.
Control by body-weight alone, which cannot recover a stall in a gust.
A person aboard before the glider has flown empty as a kite.
Gliding in gusty wind, or from a steep slope.
How you know it worked
VerificationModels built to the design glide straight and recover from a push; the lift and drag of the wing section are measured and recorded; the full-size glider has flown empty on lines and answered every control from the ground; and nobody has flown it without training on a proven design.