Water-source VRF modeling using current available curve set (as of October 2023)
IES have uncovered technical issues with the current set of water-source VRF performance curves, as provided by LG. These are presently the only set of curves available for modeling water-source VRF in ApacheHVAC (i.e., in which multiple VRF systems benefit from heat recovery, acquisition, and rejection via a common source/sink water loop or Heat Transfer Loop). As the results using these performance curves are in many cases, if not most, very reasonable, IES have opted to leave the curves in place until replacement curves are provided by LG or another VRF equipment manufacturer. Users should, however, be aware of the inconsistencies described below, and should scrutinize results accordingly.
The issue is consistency of the curve evaluation results under particular circumstances that, unfortunately, defy clear characterization even after extensive testing.
- Water-source VRF has the potential to simultaneously benefit from heat recovery, optimal part-load operation, and the more or less ideal "outdoor" unit conditions for condensing operation provided by the common source/sink water loop. Thus, according to the engineers at LG, peak COP values in the range of 20-30 are reasonable for water-source VRF under such ideal conditions.
- In many cases, perhaps most often, the results using the current set of curves from LG are both reasonable and consistent with what LG has said we ought to expect (see point above).
- In a moderate number of cases, while most parameters such as operating capacity still appear normal, the COP is notably above what would be considered reasonable, even with the ideal conditions described above. Based on the information provided by LG, any COP values greater than about 20-30 would be considered unreasonable. Professional judgement should be exercised, and this may necessitate checking coincident parameters to determine if those parameters and the COP values are appropriate.
- In a small number of cases, the current set of water-source VRF curves from LG return nonsensical results, including extremely low operating capacities resulting in extreme high part-load fraction and coincident with extreme high COP (e.g., COP reported as frequently pegged at 100, which is simply the highest COP value that can be recorded in the results file). In such cases, the problems with the coefficients on a number of the curves and the nature of the particular model and operating conditions tend to produce rather obvious and frequent departures from realistic results. In these cases, the current set of water-source VRF curves from LG are clearly not valid and should not be used.
Given the particulars of the anomalies with this curve set from LG and that IES does not have the data from which the curves were generated, it is not feasible to correct the existing curve coefficients or modify or constrain the results of the curve evaluations without invalidating the model. The current set of water-source VRF performance curves from LG will therefore be replaced as soon as a new set of curves can be obtained.