UL 2182 2017 pdf download – Refrigerants.
5.1.4 To simulate vapor leaks from equipment, the container is to be liquid filled at ambient temperature to 15 percent of mass (of maximum permissible by US DOT CFR Title 49 or the equivalent specifications in a given country of origin) fill and then leaked, at the following temperatures: a) 60°C (140°F), b) Minus 40°C (minus 40°F) or the normal boiling point plus 1 0°C (1 8°F) or bubble point at 1 atm. plus 10°C (18°F), whichever is warmer, c) Ambient Temperature, d) If (c) produces a more flammable blend than shown in (a) or (b), a temperature between (a) and (c) or (b) and (c) that causes the worst-case fractionation should be conducted. In the fractionation experiment, the composition of the head space gas and remaining liquid shall be determined by analysis (i.e., gas chromatography). Analyses shall be made initially after 2 percent of the total charge has leaked, next at 10 percent weight loss, and then at intervals of 10 percent weight losses until the cylinder reaches atmospheric pressure or no liquid remains. In the event that liquid remains and atmospheric pressure is not reached, the last analysis shall be made at 95 percent weight loss. 5.2 Leak/Recharge testing 5.2.1 Refrigerant blends containing a flammable component(s) shall be evaluated in accordance with 5.2.2 to determine the fractionation effect of successive leakage and recharging on the composition of the blend. 5.2.2 A cylinder is to be charged to 15 percent (of maximum permissible by DOT or the equivalent specifications in a given country of origin) fill with the worst-case formulation. A vapor leak at a rate of 2 percent by weight of the starting charge per hour from the vapor phase of the cylinder is to be created and maintained at ambient temperature [23°C (73°F)] until 20 percent of the starting charge has leaked. The cylinder is to be again charged with the worst-case formulation to 15 percent full and caused to leak in the same manner.
5.3 Fractionation determination by computer modeling 5.3.1 Computer or mathematical modeling may be used to obtain worst-case fractionation composition (WCFF) data. The model accuracy shall be verified if used to simulate the leakage and the leak/recharge testing conditions described in 5.1 and 5.2. 5.3.2 The fractionation composition(s) determined by modeling to be the WCFF(s) shall be confirmed by experimental verification. Experimental verification of the model shall take the form of leakage experiments (carried out in accordance with 5.1) that result in a WCFF. For blends of three or fewer components where the initial composition of the vapor or liquid phase results in a WCFF, the verification may be experimental vapor-liquid equilibrium data (VLE) at the temperature of the WCFF or over a range of temperatures that includes the temperature of the WCFF; such experiments may be carried out by performing the required tests described in 5.1 or VLE data may be taken from peer-reviewed literature.