And in fact, it’s taken a heck of a lot of evolution to get to the point where even what we consider “normal” atmospheric oxygen levels aren’t deadly.
One of the big causes was the Velcro fuzz that had been put all over the inside of the capsule, handy for sticking weightless things with Velcro hooks on them. In ordinary atmosphere, even exposed to flame it just smoulders and goes out a second later when the flame is removed. In pure O[sub]2[/sub] at 19 PSI (one atmosphere plus 5 PSI) instead of the 5 PSI they were going to run during a real mission, just the spark from a worn wire that had been fed through the hatch was enough for it to burst into flame and keep burning vigorously.
I think ** mixdenny** is talking about industrial oxygen. When you design an oxygen piping system at higher pressures, you have to careful about the elbows (bends) on the pipes because oxygen impinging on the metal bends (often stainless steel)can cause them to burn. It’s usual practice to use ultra long radius bend (the radius of curvature of the bends are 5 times the pipe diameter or more) to provide a gradual change in direction to the oxygen flowing inside.
Similarly ball valves I.e.the most common type of valves cannot be used in HP oxygen service since opening these valves a tiny bit provides a gushing flow of oxygen that is enough to set the pipe and consequently the valve on fire.
Also, in the chemical industry, there are no compressors that work on HP oxygen. Instead liquid oxygen is pumped to high pressures and then vaporized to give you high pressure gaseous oxygen. Try to compress oxygen at high pressures with most known compressors and m your compressor will burn whether it is a reciprocating or centrifugal. For contrast, Industrial compressors for hydrogen are common place even for high pressures like those used for ammonia.
The only other thing I can think of is liquid explosives that are manufactured to flow by gravity so that it is manufactured at the top of a hill and flows down for processing in the valleys below. And there are explosion catchers / relief spaces in the pipes leading down the hill.
am77494. Thanks for the nightmare fuel (if you’ll pardon the expression). 
Yes, exactly. We were using 3000 psi LOX flowing at insane levels to very high pressure rocket engines with the combustion process in the chamber at 2000 psi. We’re talking stainless steel pipes 4" OD with 1" thick walls that burn like a Roman candle in an engine failure.
The most burn resistant metal was Inconel and it was used in the most critical areas. The rocket motors themselves were made of nickel, a special alloy named NASA Z. The internal passages in the motors were so complex that they could not be made using any form of machining. A plastic model of the fuel and oxidizer channels were covered with pure carbon and electroplated to solid, over 1 inch thick at points.
This took weeks of time in the plating baths. Every so often they were removed and rough machined on the outside until they were thick enough. I really can’t do justice to the shapes of the passages. There were dozens of them running from the chamber face down the length of the skirt and back. They started out wide and not very tall and slimmed in one dimension as they increased in the other so that by the time they passed by the throat (tightest spot on the skirt) they were very tall and narrow, just the opposite of what they started out as. Then they transitioned back in their return path.
There is nothing that can stand up to the burn of 2000 psi combustion chambers. The only way to make them live is to remove the heat as fast as it enters the nickel jacket. This requires walls as thin as 0.020" which could never withstand 2000 psi. So the pressure of the oxygen in the passages had to equal the combustion pressure. The inside of the nozzles were plated with pure gold to reflect as much heat as possible.
We started and restarted every engine to failure. Sooner or later, KABOOM! I was only 30 feet away in the termination room (aptly named!) looking through 40 panes of bulletproof glass with my hand poised over the abort button. Ahh, the good old days…
Dennis