For any transport-category airplane with more than 44 seats, the FAA requires proof that a full cabin can reach the ground in 90 seconds. The test load isn’t random: at least 40 percent female, 35 percent over 50, and no more than half the exits usable.
The wing-bend test is just as real but far less public. Regulations require wings to survive 150 percent of limit load for three seconds — yet no public document spells out the hydraulic loading procedure or a pass/fail damage threshold.
The surprising thing about aircraft certification isn’t that regulators bend a wing until it proves itself. It’s that one of the most physically dramatic tests in aviation leaves almost no public paper trail around how it’s actually run.
Boeing’s 787 wingtips rose about 25 feet under ultimate load. Airbus’s A350 deflected more than five meters. Those numbers come from press releases.
Try to find the hydraulic load schedule or the damage threshold, and the trail stops.
The 90-second evacuation demonstration is the opposite. Nearly every parameter is written in binding regulation. That split — one test governed by hard public numbers, the other real but largely undocumented — is the actual story.
The 90-second rule is built on numbers that don’t bend
The trigger is precise. Under 14 CFR 25.803(c), any transport-category airplane built for more than 44 passenger seats must prove, in a simulated emergency, that its full seating capacity plus required crew can get to the ground in 90 seconds.
That sentence does a lot of work. It creates a pass-or-fail clock, and it applies regardless of how large or innovative the cabin is.
The appendix J criteria turn the passenger load into a demographic specification: at least 40 percent female, at least 35 percent over 50, and at least 15 percent both female and over 50.
In other words, the test pool is structured to resemble the traveling public, not a group of fit volunteers.
A realistic passenger mix also means slower movement. That’s the point.
Then there’s the exit limit. A full-scale evacuation demonstration may use no more than 50 percent of the airplane’s emergency exits. Unused exits have to be marked unusable and chosen to represent the full exit layout.
The aircraft must prove it can empty even when half the doors are out of action.
Configuration is fixed too: the airplane sits in a normal ground attitude with landing gear extended and all emergency electrical power sources deenergized. Lighting is effectively emergency-only.
Operators face their own version. The current text of 14 CFR 121.291, hosted by the Legal Information Institute, requires air carriers to show the same full-capacity evacuation in 90 seconds or less. They also run a partial evacuation demonstration focused on crew: flight attendants must open half of the required floor-level exits and half of the required non-floor-level exits, deploy half the slides, and have those exits and slides ready within 15 seconds.
| Requirement | Applies to | Key numeric/condition parameters | Primary regulatory or documented source |
|---|---|---|---|
| Trigger for full-scale evacuation demonstration | Transport-category airplanes with seating capacity greater than 44 passengers | Maximum seating capacity including required crew can evacuate to ground within 90 seconds under simulated emergency conditions | 14 CFR 25.803(c) |
| 90-second evacuation time limit | Manufacturer and air-carrier evacuation demonstrations for airplanes with more than 44 passenger seats | All occupants must reach ground within 90 seconds; air carriers must show full capacity can evacuate in 90 seconds or less | 14 CFR 25.803(c); 14 CFR 121.291(a) |
| Passenger mix in full-scale evacuation test | Full-scale evacuation demonstrations under appendix J for part 25 certification | Representative load: ≥40% female; ≥35% over age 50; ≥15% both | 14 CFR part 25, appendix J(h) |
| Maximum exits usable in certification evacuation | Emergency evacuation demonstration for airplanes subject to part 25 | No more than 50% of emergency exits; unused exits marked unusable and representative | 14 CFR part 25, appendix J |
| Aircraft configuration during evacuation test | Part 25 full-scale evacuation demonstration | Normal ground attitude, landing gear extended, emergency power off; lighting emergency-only | 14 CFR part 25, appendix J |
| Crew slide-opening proficiency (partial evacuation) | Air-carrier partial evacuation demonstration | Flight attendants must open 50% floor-level and 50% non-floor-level exits, deploy 50% slides; ready in 15 seconds | 14 CFR 121.291(c)(1) |
| 1965 two-minute precursor standard | Part 121 air-carrier evacuation demonstrations before 1967 | Initial maximum 120 seconds; later reduced to 90 seconds as slide tech improved | OTA Aircraft Evacuation Testing report; FAA Amendment 25-15 |
| Analysis/testing substitution for full-scale demonstration | Part 25 emergency evacuation compliance options | Administrator may accept analysis combined with testing when equivalent to full-scale test | 14 CFR 25.803(c) |
| Static wing-bend strength requirement | Transport-category airplane wings and primary structure | Support limit loads without permanent deformation; ultimate loads at 1.5× limit for 3 seconds; public text lacks rig details | 14 CFR 25.307; 25.305; AC 25.307-1 |
| Documented public wing-bend demonstrations | Boeing 787, Airbus A350 XWB | 787 wing flex 25 ft at 150% loads; A350 1.5× limit, >5 m deflection, successful | Boeing press materials; Aviation Week coverage |
The wing-bend test: real on paper, thin on documentation
Structurally, the rule is unambiguous. Under 14 CFR 25.307, the airframe must carry limit loads without harmful permanent deformation. It then has to withstand ultimate loads — 150 percent of limit — and remain intact for at least three seconds.
What the regulation doesn’t say is what a test engineer would most want.
No dedicated wing-bend rig setup. No hydraulic loading scheme. No explicit pass/fail damage threshold.
A search across FAA structural regulations, advisory circulars, and major manufacturer announcements found only broad proof-of-structure language.
There is physical evidence that the tests get run.
Boeing flexed the 787’s static airframe at Everett on March 28, 2010. The wingtips rose about 25 feet while the structure held 150 percent of the most extreme in-service loads. Airbus reported the A350 XWB wing loaded to 1.5 times limit load in December 2013, deflecting more than five meters at the tip.
Yet neither manufacturer disclosed its hydraulic loading setup or acceptable-damage limits. Certification history shows no case in which a failed static wing test, by itself, altered or blocked a certification decision.
Ask a certification engineer about the wing-bend test and you’ll hear confidence. Ask for the hydraulic load schedule, and the conversation quiets down.
Why the 90 seconds exists in the first place
The two-minute clock came first. In 1965, under Amendment 121-2, U.S. air carriers had to demonstrate that every passenger and crewmember could get out in 120 seconds, according to an Office of Technology Assessment report hosted by Princeton University.
Amendment 25-15 followed in 1967, adding evacuation rules to part 25 and shortening the clock to 90 seconds. The FAA tied the tighter limit to advances in automatic slide deployment and to research on post-crash fire survivability. Once slides popped automatically, the shorter time was no longer hostage to crew speed alone.
Today’s rule still hangs on that logic: 14 CFR 25.803(c) is the anchor, and appendix J — not a policy memo — converts the demographic mix and exit cap into binding certification criteria.
Questions? Answers.
What is the 90 second rule in aviation?
This is an FAA certification standard for airplanes designed to carry more than 44 passengers. In a simulated emergency, every passenger seat plus the required crew must clear to the ground within 90 seconds. In practice, that means a full-scale demonstration or an approved combination of analysis and testing.
How many seconds does it take to evacuate a plane?
Certification requires 90 seconds or less. The 2006 A380 demonstration evacuated 873 people in about 78 seconds, but that was a planned test environment.
Why must evacuation certification tests include a minimum share of women and passengers older than 50?
Appendix J requires at least 40% female and 35% over 50, including 15% both, so the test isn’t stacked with younger athletes. A realistic demographic mix better reflects who will actually be on board.
Why can only half of an airplane’s emergency exits be used during a full-scale evacuation demonstration?
It assumes some exits will be blocked by fire or damage. If occupants can get out using no more than half the exits within 90 seconds, the aircraft meets a more conservative safety case.
Why is detailed wing-bend test procedure hard to find in public documents?
Public FAA and EASA texts specify ultimate loads at 150 percent of limit loads and a three-second no-failure requirement, but not hydraulic loading procedures or pass/fail damage thresholds. The research found no current public document with those engineering specifics.
Flight deals
most people never see
Our AI monitors 150+ airlines for pricing anomalies that traditional search engines miss. Air Traveler Club members save $650 per trip per person on average: see how it works.
Each deal saves 40–80% vs. regular fares:
