Oxygen service is unusual on this site because almost nothing about the metal changes. The grade is the same, the corrugation is the same, the braid is the same. What changes is how clean the inside of the assembly has to be, how that cleanliness is achieved and verified, and who is willing to put their name to it.
What this service does to a hose
Oxygen does not attack stainless steel. It makes everything else in the bore dangerous. A film of drawing lubricant, a fingerprint, a trace of cutting oil left from machining a ferrule, a fibre from a cloth — in an oxygen-enriched atmosphere any of these will ignite at a temperature it would never reach in air, and once it does the metal can sustain the burn.
The ignition source is usually not a spark. It is adiabatic compression: opening a valve quickly onto a dead-ended line compresses the gas in front of the flow, and that gas gets hot — hot enough, in a fast opening, to light a contaminant. This is why oxygen practice cares so much about how fast valves are opened, and why velocity limits on oxygen lines are tighter than the flow itself would demand.
Particulate matters for the same reason. A loose particle travelling at velocity carries kinetic energy, and where it strikes an impingement point it deposits that energy in one spot.
What we would fit
| Element | What we would supply | Why |
|---|---|---|
| Hose | 316L annular corrugated, degreased and cleaned for oxygen service | The grade is ordinary; the cleaning is not |
| Braid | 316L, sized conventionally | It is outside the bore and does not see the medium |
| Liner | Only where velocity demands it, and cleaned to the same standard as the bore | A liner adds surface area that also has to be clean |
| Ends | Metal-to-metal or with oxygen-compatible seals — specify the seal material | Elastomers and greases are the usual contaminant, not the steel |
| Packaging | Capped and double-bagged immediately after cleaning, and not opened until fitting | An assembly cleaned and then left open on a bench is no longer clean |
What we would not fit, and what happens
An ordinary production assembly, cleaned afterwards. Cleaning is not a finishing step you can add to a hose that has been made, handled and stored normally. It changes how the parts are handled from the ferrule onward.
Interlock hose. Not pressure-tight, and the seam is impossible to verify clean.
PTFE tape, thread sealant or any grease at the ends. If the joint needs a sealant, it is the wrong joint for this duty.
A hose sized on flow alone. Oxygen velocity limits are set by ignition risk at impingement points, not by pressure drop, and they are frequently well below what the general velocity limits would allow. Where the two disagree, the oxygen practice wins.
What we need to size it
| We need | Because |
|---|---|
| Which oxygen cleanliness standard and level applies | This is the specification; everything else is secondary to it |
| Gaseous or liquid oxygen, and the temperature | LOX is also a cryogenic duty and contraction has to be sized in |
| Working pressure and how fast the line is valved | Fast valving onto a dead end is the classic ignition case |
| Flow rate and bore, or the velocity you are held to | Oxygen velocity limits are usually tighter than ours |
| Seal and gasket materials permitted by the client spec | The seal is more often the problem than the steel |
| Whether third-party verification of cleaning is required | It changes who has to do the work and how it is documented |