The antenna that folds every time it lands
Physics says SkyFall’s radar antenna should be 48.3 centimetres long. The helicopter sits 15.2 centimetres off the ground. Cut the antenna to fit and it loses 54 percent of its radar band — the half that sees deep enough to find ice.
NASA’s next Mars aircraft is not one helicopter but three. The SkyFall mission, planned to launch in late 2028, will fly a trio of rotorcraft that follow Ingenuity, the helicopter that flew 72 times over nearly three years and proved powered flight works in Mars’s thin air. Each SkyFall craft carries four instruments. One of them is a ground-penetrating radar, and its job is to find water.
Not deep water. Orbiting spacecraft can already map thick ice deposits tens of yards down. What they are effectively blind to is the top several yards of regolith, the broken rock and dust at the surface — which happens to be the only ice future astronauts could easily dig up for water, oxygen and fuel. Seeing that layer means flying low and slow with the radar pointed at the ground. And that is where an engineering problem shows up that you can solve with one division.
A band, not a frequency
SkyFall’s radar does not transmit on one frequency. It sweeps a band from 500 to 2,500 megahertz. A radio wave’s length is set by its frequency: divide the speed of light by the frequency and you get the wavelength. At 500 MHz that is 60 centimetres. At 2,500 MHz it is 12. The top of the band is five times the bottom, so the longest wave is five times the shortest.
The two ends do different jobs. According to JPL, the longest wavelengths reach several yards below the surface, while the shortest resolve fine detail in the uppermost layers. You need both to see where dry soil gives way to ice — and an antenna has to be built for the longest wave it will ever handle.
The number that doesn’t fit
A traditional antenna covering that band, JPL says, would need to be 48.3 centimetres long, with a clear view of the ground. The gap between the Martian surface and the bottom of a SkyFall fuselage is about 15.2 centimetres.
Fifteen point two divided by forty-eight point three is 0.31. The antenna gets less than a third of the length it wants, so it has to shrink by a factor of 3.2. Hang a traditional antenna under the craft anyway and 33 centimetres of it would be underground at touchdown.
“…it is about 1½ times longer than the helicopter’s legs.”
That is Christine Gebara, SkyFall’s ground-penetrating radar mechanical lead at JPL, describing the antenna after her team had shrunk it. Even miniaturised, it is longer than the legs it hangs between.
Why not just make it smaller?
Here is the obvious fix, and the reason it fails. A traditional antenna’s length tracks the longest wave it serves: 48.3 centimetres for a 60-centimetre wave, a ratio of about 0.805. That ratio is this site’s model, not a NASA rule, but it is built from JPL’s own two numbers, and it lets you run the problem backwards.
Cut a traditional antenna to 15.2 centimetres and the longest wave it can serve is 15.2 ÷ 0.805, about 18.9 centimetres. That is a frequency of roughly 1,590 MHz. Everything between 500 and 1,590 MHz is gone: 54 percent of the band. The radar’s range, top frequency over bottom, collapses from five to one to about 1.6 to one.
And it is the wrong 54 percent to lose. What disappears is the long-wavelength end, the part that penetrates deepest. A short rigid antenna would still image surface texture beautifully. It would miss the ice the mission exists to find.
So it folds
JPL’s answer was to change what the antenna is made of rather than how long it is. After an extensive search the team chose a Vivaldi antenna: a flat, flared design that works across a wide, continuous band and can be cut from flexible metallised fabric. (Its inventor, Peter Gibson, thought the sweeping curves looked like a violin.) Because SkyFall surveys shallow ground, under about five metres, in dry regolith that blocks radio waves far less than Earth’s soil, the team could shrink it further without giving up sensitivity.
It still doesn’t fit, so it bends. At landing the Vivaldi folds out of the way — further if the helicopter comes down on a rock — and springs back into shape for flight. It is sheathed in polyester and layers of Vectran, the material of the Spirit and Opportunity landing airbags, and held in shape by fibreglass tape springs on a magnesium mount. The whole assembly weighs about 150 grams.
That trade turns a geometry problem into a reliability problem. A rigid antenna that fits fails on the first day, quietly, by not seeing the ice. A folding antenna could fail on landing number forty. So JPL tested it: bent it, cycled it through Mars-like day–night temperature swings of up to 94 °C, flexed it as if through dozens of landings, and stopped six times along the way to check that its signal had not degraded. By the end it had survived 200 landings, more than double what the prime mission requires, with no loss of performance.
Read that last number carefully. “More than double” tells you the requirement is under 100 landings. It does not tell you what the requirement is. It is a bound, not a value, and the difference matters every time a press release reports a margin.
The antenna is not finished. It has cleared its first big test campaign and is not yet flight-qualified: an engineering model still faces vibration testing, deployment in a simulated Martian environment, more signal tests and trials at JPL’s Mars Yard. The launch date will not move to wait for it. How much testing is enough is the one question in this story without an answer key.
Try it
Size the antenna
Pick a frequency and a ground clearance. The sizer turns frequency into wavelength, applies this article’s sizing ratio, and tells you whether a rigid antenna fits — and how much of SkyFall’s 500–2,500 MHz band a rigid antenna that fits could still reach.
SkyFall’s band runs 500 to 2,500
SkyFall: about 15.2. Try 8, or 48.3.
Wavelength = speed of light ÷ frequency
| Quantity | Value |
|---|---|
| Rigid antenna needed (model, 0.805 × wavelength) | 48.3 cm |
| Room under the craft | 15.2 cm |
| Does it fit? | No |
| Lowest frequency a rigid antenna that fits can reach | 1,589 MHz |
| Half-wave dipole, length of each arm | 15.0 cm |
| Share of SkyFall’s band still reachable | 46% |
Model, not NASA’s rule. The 0.805 ratio is built from JPL’s two published numbers, 48.3 cm for a 60 cm wave; real antennas do not all follow one ratio. The dipole arm is the textbook quarter wavelength; real arms are usually trimmed a few percent shorter.
Classroom packet · 22 pages
The Antenna That Had to Fold — Aerospace Engineering Case Study
The full classroom version of this analysis, built as a 90-minute block for grades 10–12 and higher education, with a two-period split and a 50-minute short version. Print-ready PDF. Calculators only — no lab, no software, no prep.
- Students turn the 500–2,500 MHz band into wavelengths and derive the sizing ratio the source never states.
- Part 2B is the calculation above — under a third of the space, and a 54 percent band loss for the obvious fix — done from scratch.
- Part 3 runs it backwards for an invented 8 cm helicopter: a rigid antenna that fits reaches nothing in the band at all.
- Part 4 ends on a decision with no answer key — commit the antenna or keep testing — graded on a four-criterion rubric. Full answer key with every step worked.
Left: the classroom packet on Teachers Pay Teachers. Right: a USB software-defined radio receiver that tunes up to 1.7 GHz and ships with a dipole antenna kit in two lengths — because, as this article shows, an antenna’s size is set by the wavelength it has to catch — an Amazon affiliate link. As an Amazon Associate this site earns from qualifying purchases, at no additional cost to you.
Sources
NASA/JPL, “NASA Tests Featherweight Radar Antenna for SkyFall Mars Helicopters,” news release 2026-055, August 6, 2026 — source of every measured figure: the 500–2,500 MHz band, the 60 and 12 cm wavelengths, the 48.3 cm antenna, the 15.2 cm clearance, the 1½× leg ratio, the 200 landings, the 150 g mass and the 94 °C swing. Universe Today, “NASA’s SkyFall Mars Helicopters Will Feature a Revolutionary Antenna Design” (secondary). Modelled, not measured: the 0.805 sizing ratio, the 1,590 MHz cutoff, the 54 percent band loss and the sizer above are this site’s calculations from JPL’s numbers, not NASA figures.