There are various methods for sampling nitrox, however,
the method I prefer (as it's matched to the adaptor design) is as follows:-
1. Attach the adaptor to the cylinder
2. Plug in and calibrate the oxygen sensor.
3. Open the cylinder until a faint hiss is heard
4. Push fit the oxygen sensor into the adaptor
5. Wait until a steady reading is obtained on the display
6. Turn off the cylinder
7. Wait until a steady reading is obtained on the display (about 0.5 -
1.5% LESS than in 5.)
8. Repeat (or until a consistent result for 7. is obtained)
OK, why the difference in readings in 5 and 7?
This is caused by back pressure. With gas flowing through
the system, the pressure in the adaptor will be greater than ambient.
Seeing as the oxygen sensor works on partial pressures of oxygen,
the reading will be artificially high (1). When the gas flow from the
cylinder is turned off, the pressure inside the adaptor will quickly fall
to ambient due to the exhaust. However, the gas will not vent off too
quickly because of the exhaust, with the reading staying constant for
at least 30 seconds or so. Using this method, I can reliably test gas
mixes to a maximum of 0.3% deviation from shop analysers.
(1) Does that mean shop readings are high?
Depends on the analyser. When Neil in Bristol
Channel Diving tests nitrox mixes, he uses a constant flow of around
4l/min. However, before testing the mix, the unit is calibrated in the
same constant flow of air. Therefore the readings are comparable. Golden
rule here is.... calibrate at ambient, test at ambient.