Chiller Purge Pump-Out Leak Check

It depends on the size of the machine, and Trane publishes the answer as a chart rather than a number. Trane CTV-SVX05H-EN (August 2020), page 7, Figure 2 plots purge minutes per day against chiller tons per circuit from 400 to 2000 tons in three bands: Typical Operation, Small Leak and Large Leak. Read off the figure, typical operation runs up to about 4 minutes a day at 400 tons and about 20 at 2000 tons; a 400-ton circuit is into Trane's small-leak band above 4 and its large-leak band above 28, while a 2000-ton circuit has room to 20 and 60. A single figure such as five minutes a day is the chart read at 500 tons and is wrong at the others; fifteen minutes a week is about two a day, inside typical operation at every tonnage Trane plots, which is why it can sound right to one operator and tell another nothing. The chart is for R-123 and R-514A CenTraVac chillers; it is not a rule for other machines.

No. Trane's own caption says the figure "was created using the average annual refrigerant leakage rate to be calculated based on the daily purge pump out time and the unit refrigerant charge", and prints none of the constants that would let you run that backwards. So the chart tells you Trane's band, and that is all this tool reports. The leak repair rules (40 CFR 84.106 for HFC appliances, 82.157 for ozone-depleting refrigerants) are applied to the refrigerant you add back, on every addition. The leak-rate calculator link on this page carries the chiller's refrigerant and appliance type so the engine can do that; it does not carry an addition, so the calculator asks for the pounds you added.

R-123: Ozone-depleting substance (HCFC-123, class II under 40 CFR part 82 subpart A), excluded by 84.106(a)(3)(i). R-123 chillers stay under the Section 608 rule, 40 CFR 82.157, which reaches them at a full charge of 50 lb or more with the same 10 percent comfort-cooling trigger. R-514A: Non-HFC substitute with GWP 3 as EPA prints it in the Technology Transitions final rule; its constituents HFO-1336mzz(Z) (GWP 2) and HCO-1130(E) (GWP 5) at 74.7/25.3 percent by mass are both in Table 1 to 84.64(b), the table 84.106(a)(2) names, and mass-weight to 2.8; EPA's SNAP chiller listing prints 7. At or below the 53 cutoff at any of them, not covered. Neither constituent is a class I or class II ozone-depleting substance, so 82.157 does not reach it either. Both sentences are the leak-rate engine's own, and its calculator page applies the matching rule to your additions.

Because Figure 2's x-axis is "Chiller Tons per Circuit", not the nameplate tonnage of the whole machine. Read your controller's pump-out figure against the tonnage of the circuit that purge serves: a 1000-ton circuit has a typical-operation ceiling of 10 minutes a day whatever the machine around it adds up to. Trane's handling guide does not print how purges are arranged on a given model; your unit's own service literature does.

Trane plots 9 tonnages, 400 to 2000 in steps of 200. Between them the check reads the band ceilings linearly and says so on the result: a 500-ton circuit gets a typical-operation ceiling of 5, a small-leak ceiling of 30 and a large-leak ceiling of 70.5 minutes a day. Outside 400 to 2000 tons the check refuses rather than extrapolate a chart Trane did not draw. The two independent readings of the printed figure agree within 1 minute per day, and a reading that close to a boundary is flagged as being fairly readable either way.

Not as printed. The chart is Trane's, for Trane's CenTraVac machines running R-123 or R-514A, and it says so on its cover. Another manufacturer's purge unit has its own capacity and its own factory recommended values, and the check does not guess them. What does carry over is the shape of the question: the acceptable pump-out time is a function of machine size, so any single scalar rule of thumb is wrong for most of the machines it is applied to.