Nitrox 31% Instead of 32%: A Safety Problem or a Normal Reality?

The number on the label isn't the problem — your MOD is

A mix's oxygen percentage only matters because it sets two things: your no-decompression limit (NDL) and your maximum operating depth, the MOD.

The MOD is calculated from the partial pressure of oxygen (PpO2). The standard working limit is 1.4 bar. The 1.6 bar limit is generally treated as a maximum (contingency) limit in recreational diving, not a comfort zone. Formula: MOD (m) = (max PpO2 ÷ oxygen fraction − 1) × 10.

The direct, counter-intuitive consequence: a weaker oxygen mix gives you a deeper MOD, so more oxygen margin. Nitrox at 31% isn't "less safe" than 32%. On the oxygen side, it's actually a touch more forgiving. You only lose a small no-decompression benefit.

Working limit

PpO2 1.4 bar

Contingency limit

PpO2 1.6 bar

Mix variation

A few tenths

Analyzer-to-analyzer spread

Up to 2–3 points

EAN31, EAN32, EAN33: what it actually changes underwater

Here's how a few common mixes translate at PpO2 1.4 bar.

One percentage point shifts the MOD by about a metre and changes the no-deco time at the margin. None of this is dangerous as long as you analyze and set your computer to the real value.

Why a Nitrox tank is never a "to-the-decimal" product

An enriched-air mix is made, not poured. Whether by continuous blending (oxygen is injected into the air stream before compression) or by partial pressures (you load oxygen first, then top up with air), the result aims for a target range, not an exact decimal.

Pressure, temperature, humidity, flow rate, the day's setup: all of it shifts the result by a few tenths. Aiming for 32% and getting 31.3% or 32.4% is perfectly normal. In practice, some variation around the target is expected. That's exactly why every diver analyzes their own tank and sets their computer to the value actually read.

Demanding "exactly 32.0%" is like asking a baker for a loaf that weighs 500.0 g to the gram. That's not how gas physics works.

Analyzer, flow, temperature: field measurement isn't a laboratory

The percentage you read also depends on the analyzer. Most use a galvanic cell, calibrated on ambient air at 20.9% before each session. That cell ages, reacts to temperature, and the gas flow at the tank outlet influences the reading.

The consequence: a gap between 31% and 32% can simply fall within the device's measurement uncertainty. No analyzer is "lying". That's why the rule isn't "have the right number", it's "analyze it yourself and set your computer to what you read".

When three analyzers give three readings: the spread can reach 2 to 3 points

Here's the real scene, the one twenty years on the water confirm every single day. On the same boat, on the same tank, three analyzers of different brands almost never give the same number. With ours — a DE-OX, a DNA and an NCR — we routinely see 2 to 3 points of spread between devices, while the Bauer membrane itself fills steadily around 32–33%.

In practice, you can read this on the same tank:

When the spread reaches 2 or 3 points, we can't call it "a few tenths" anymore. But it doesn't mean the tank is bad either. What varies is the whole measurement: the brand, the galvanic cell, its age, its calibration, the flow, the temperature, the humidity, the protocol and the conditions on deck.

A Nitrox analyzer isn't a NASA or MIT instrument in a clean room. It's a field tool, used on a boat, in the heat, salt and humidity, sometimes with an ageing cell. Two different brands, two cells of different age: the spread is expected, not abnormal.

A 2-to-3-point spread between analyzers doesn't immediately prove the gas is bad. First, it proves the measurement must be checked properly.

Analyzer spread: the Dragon Dive Komodo rule

When the analyzers disagree, we don't settle it by opinion or ego. We go back to procedure, and we treat the spread according to its size.

Small spread — 0 to 1 point

This is the normal reality of fieldwork. We note the value read on a properly calibrated analyzer, set the computer to it, respect the MOD, and dive.

Moderate spread — 1 to 2 points

We redo the analysis: same flow, same sampling point, recalibrated analyzer, stabilized reading. We compare against the centre's reference analyzer, then settle on a consistent value.

Large spread — 2 to 3 points or more

We don't hold a debate on the boat deck. We treat it as a disagreement between instruments. A tank is never validated on a single isolated analyzer: we check against the Dragon Dive reference analyzer, control the flow, redo the calibration. If one analyzer stays consistently offset, it's considered unreliable for that analysis — the problem isn't the Nitrox, it's the instrument.

1. Calibrate the analyzer on air (20.9%), under the current conditions.
2. Slow, steady flow, with no over-pressure directly on the cell.
3. A stabilized reading, never a value grabbed in one second.
4. Compare against the Dragon Dive reference analyzer.
5. If the spread stays large, the inconsistent analyzer is set aside or the tank is rechecked.

And the point that's never up for negotiation: if the spread doesn't resolve, you don't pick the "comfortable" number. You plan conservatively, or you recheck the tank. We don't dive on a marketing number. We dive on a validated analysis.

The field reflex: the analyzer's value, and never deeper than the guides

Here's the most telling case. A tank blended around 32% that reads 29% on the analyzer aboard the boat. What do we do?

We always keep the analyzer's value. We set the computer to 29%, never to the fill-station number. And we never exceed the depth set by the guides. That simple-looking reflex stacks three layers of safety.

Three margins stacking on top of each other. That's exactly what field safety is: not a perfect number on a screen, but layers that add up. Nitrox safety doesn't come from a tidy round figure. It comes from correct procedure and respected margins.

What a Nitrox station in Komodo really looks like

At Dragon Dive Komodo, the gas is produced on site. Four electric Bauer compressors feed the station, with Bauer Securus filtration monitoring — the only setup of its kind at a dive resort in Labuan Bajo. Every tank is filled in-house, never outsourced.

In concrete terms, that means direct control over air quality, over the Nitrox, and over what goes into your tank. No opaque fills, no label slapped on without analysis. You analyze your tank, you note the value, and you leave with a verified figure — whether it reads 31, 32 or 33.

A well-controlled gas setup lets us adapt the equipment to the diver, not the other way round.

The real safety rule: analyze, note, set, respect

The right reflex is never to demand "exactly 32.0%". The right reflex comes down to five steps, identical for every Nitrox diver in the world:

1. Analyze your tank before every dive.
2. Note the actual percentage read on the analyzer.
3. Calculate or check the matching MOD.
4. Set your computer to the actual percentage, not the one on the label.
5. Respect the maximum depth.

A diver who applies these five steps is safe with 31%, 32% or 33%. A diver who ignores them can be at risk even with a "perfect" mix.

Why Nitrox is still excellent for diving Komodo

The point of Nitrox isn't marketing comfort. It's one precise, proven thing: less nitrogen in the mix, so a lower nitrogen load at the same depth. That translates into longer no-decompression times and more efficient surface intervals.

In Komodo, where you often string several dives together in a day on repetitive profiles, this is exactly the terrain where Nitrox earns its place: more margin between dives, more flexible profiles, and more useful bottom time on the sites that allow it.