Ask how long a pilot can fly without oxygen and you get two answers that do not resemble each other.

The legal answer lives in 14 CFR 91.211, and at 12,000 feet it is remarkably relaxed: fly all day. The physiological answer comes from the FAA’s own aeromedical guidance, and at 35,000 feet it is thirty seconds. The distance between those two numbers is where almost every hypoxia accident happens.

What follows is the regulation as written rather than as summarised, alongside the FAA’s published time-of-useful-consciousness figures. Both are less intuitive than the internet’s version of them.

What 14 CFR 91.211 actually says

The general rule is three sentences long and worth reading in the original, because almost every secondary summary compresses it wrongly.

No person may operate a civil aircraft of U.S. registry—

(1) At cabin pressure altitudes above 12,500 feet (MSL) up to and including 14,000 feet (MSL) unless the required minimum flight crew is provided with and uses supplemental oxygen for that part of the flight at those altitudes that is of more than 30 minutes duration;

(2) At cabin pressure altitudes above 14,000 feet (MSL) unless the required minimum flight crew is provided with and uses supplemental oxygen during the entire flight time at those altitudes; and

(3) At cabin pressure altitudes above 15,000 feet (MSL) unless each occupant of the aircraft is provided with supplemental oxygen.

Three things in that text are routinely lost in translation.

It is cabin pressure altitude, not flight altitude. A pressurised business jet cruising at FL410 with the cabin held at 7,000 feet triggers none of paragraph (a). The airframe is at eight miles; the occupants are, as far as this rule is concerned, in the foothills. Paragraph (b) is what catches pressurised aeroplanes, and it is a separate set of requirements.

It is MSL, not AGL. The rule does not care how close the ground is. A pilot crossing the Rockies at 13,000 feet is inside the first band in exactly the way a pilot at 13,000 feet over the Gulf Coast is, even though one of them can nearly touch the terrain.

The verbs are not the same in all three paragraphs. In (1) and (2) the crew is “provided with and uses” oxygen. In (3), every occupant is only “provided with” it. Above 15,000 feet your passengers must have oxygen available to them. Nothing in Part 91 obliges them to breathe it.

And the 30-minute clause is not a 30-minute allowance. It applies to the band from above 12,500 up to and including 14,000 feet, and it reads “for that part of the flight at those altitudes that is of more than 30 minutes duration” — a description of which flights the requirement attaches to, not a free half-hour. Above 14,000 feet there is no grace period at all.

Destin Sandlin’s altitude-chamber run is the most-watched hypoxia demonstration on the internet, and it earns the eleven million views: a numerate, articulate adult loses the ability to be either, while insisting he is fine.

The altitude the regulation ignores

Here is the awkward part. The FAA’s Aeronautical Information Manual, paragraph 8-1-2, does not agree with the impression 91.211 leaves.

The AIM notes that night vision starts to deteriorate at a cabin pressure altitude as low as 5,000 feet. It then describes the band from 12,000 to 15,000 feet in language a regulator does not use lightly: “judgment, memory, alertness, coordination and ability to make calculations are impaired, and headache, drowsiness, dizziness and either a sense of well‐being (euphoria) or belligerence occur.” A few lines later it adds that pilot performance can seriously deteriorate within 15 minutes at 15,000 feet.

So at 12,000 feet the regulation asks nothing of you and the information manual says your judgment is already going. Both documents are official and both are correct. 91.211 sets an enforceable floor; it was never a statement about where the physiology begins. Treating the regulation as medical advice is the first mistake in the sequence.

Time of useful consciousness: the clock that matters

The number pilots actually need is not in the regulation. It is in Advisory Circular 61-107B, which defines time of useful consciousness as “the period of time from interruption of the oxygen supply, or exposure to an oxygen-poor environment, to the time when an individual is no longer capable of taking proper corrective and protective action.”

Figure 2-3 of that circular is reproduced below. The second column is the one that should change how you think about a cruise altitude.

Two rules of thumb sit underneath that table. Between 25,000 and 43,000 feet, the circular says the time of useful consciousness following a decompression can be assumed to be cut in half. Above 43,000 feet it stops being a duration in any meaningful sense: the circular reduces it to “the time it takes for the blood to circulate from the lung to the brain, plus any reserve oxygen stored in the brain,” approximately 9 to 12 seconds.

And the circular prints one warning in bold that is easy to misread the table without: “The TUC does not mean the onset of unconsciousness. Impaired performance may be immediate.” The 30 to 60 seconds at 35,000 feet is not 30 to 60 seconds of competence. It is the window before you can no longer help yourself, and the useful part of it is shorter than the number.

The card test, filmed inside a chamber. Students are asked to name the suit and value of a playing card and describe their own symptoms out loud. Watch how long the answer to a four-of-spades takes.

The slow leak is the dangerous one

Aviation folklore treats decompression as an event: a bang, fog in the cabin, masks dropping. AC 61-107B takes the opposite view, and says so twice in the same two pages.

“Slow decompression is as dangerous as or more dangerous than a rapid or explosive decompression,” reads the first warning. The reasoning follows a paragraph later: “By its nature, a rapid decompression commands attention. In contrast, a slow decompression may go unnoticed and the resultant hypoxia may be unrecognized by the pilot.”

That sentence is the whole of the ghost-flight genre in one line. It is what happened to Helios Airways Flight 522 and, in a smaller aeroplane, to Payne Stewart’s Learjet. Neither crew was fighting a hurricane in the cabin. Both were quietly, unknowingly, losing the argument.

Above FL250: the pressurised-cabin rules

Paragraph (b) of 91.211 is where the airline and business-jet world lives, and it is built entirely around the assumption that the pressurisation will one day fail.

Above FL250, a pressurised aircraft must carry at least a 10-minute supply of supplemental oxygen for each occupant, over and above anything paragraph (a) requires, specifically “for use in the event that a descent is necessitated by loss of cabin pressurization.” Ten minutes is not a comfort margin. It is a descent budget.

Above FL350 the rule reaches into the cockpit: one pilot at the controls must be wearing and using a secured, sealed oxygen mask. There is a single relaxation, and it is the most quoted line in the section. That pilot need not wear the mask at or below FL410 if there are two pilots at the controls and each has “a quick-donning type of oxygen mask that can be placed on the face with one hand from the ready position within 5 seconds, supplying oxygen and properly secured and sealed.”

Five seconds, one hand. It is worth noticing that AC 61-107B’s emergency-procedure table opens with the same figure from the other direction: “DON MASK In 5 seconds or less. Check for flow. Breathe 100 percent oxygen.” The certification standard and the human procedure were written to meet in the middle.

One more clause is easy to skip and is the reason cockpits do not empty out in the forties: if for any reason a pilot has to leave the controls above FL350, the remaining pilot puts a mask on and keeps it on until the other one is back in the seat.

Where a mask stops being enough

At the bottom of the table the numbers stop behaving like durations, and the equipment changes with them. A mask delivering oxygen at ambient pressure has a ceiling, because at some point the surrounding air pressure is simply too low to push enough oxygen across. Beyond that, the crew needs pressure as well as oxygen, which means a suit.

IL U-2 Dragon Lady is the clearest working example. Its pilots fly in full-pressure suits maintained by the 9th Physiological Support Squadron at Beale Air Force Base, the same unit that runs the altitude chamber in the photographs above. Each suit is inspected on a rotating cycle against a date printed on the collar; the squadron also maintains the parachutes, the vent hoses and the portable liquid-oxygen carriers the pilots breathe from on the walk to the aircraft. It is a wardrobe department for the edge of the atmosphere.

For everyone below that, the circular’s remedy is blunt and repeated on four consecutive pages: if hypoxia is suspected, don the mask, breathe 100 percent oxygen slowly, descend. If supplemental oxygen is not available, start an emergency descent to below 10,000 feet MSL. The encouraging half of that guidance is the recovery: once 100 percent oxygen is administered, the circular says, recovery usually occurs in a matter of seconds. Hypoxia is one of the few emergencies in aviation that undoes itself the instant you fix the cause.

What the rules do not require

Flights above 25,000 feet MSL bring in a training requirement of their own, under 14 CFR 61.31(g), and AC 61-107B exists largely to describe what that ground training should contain. What it does not contain is a chamber ride.

The FAA’s Civil Aerospace Medical Institute put it plainly in a 2003 report, DOT/FAA/AM-03/10: pilots and crewmembers on flights exceeding 25,000 feet MSL are required to complete ground training in high-altitude physiology, including hypoxia training, “however, regulations do not require altitude chamber training.” That study existed because CAMI wanted to know what pilots thought about the gap.

It is a gap with a straightforward consequence. A pilot can be fully legal to operate in the forties having never once felt what their own hypoxia feels like. Everything in the AIM and the advisory circular says that personal symptom set is the single most useful thing a high-altitude pilot can own, and nothing in the regulations makes anyone go and get it.

A pilot declaring an emergency while hypoxic. The recording has been used in training for years for one reason: the voice sounds almost normal, and the thinking is already gone.

The regulation gives you altitudes. The advisory circular gives you seconds. Only one of the two is written on the assumption that you will be awake to read it.

Sources: 14 CFR 91.211 (eCFR); FAA Aeronautical Information Manual, paragraph 8-1-2; FAA Advisory Circular 61-107B, Chapter 2 and Figure 2-3; FAA Civil Aerospace Medical Institute report DOT/FAA/AM-03/10; Auburn University College of Liberal Arts; DVIDS.