Residential air-conditioning systems are considered essential in many parts of the United States. These products should be selected based on a comparison of the estimated heat gain to the manufacturer's performance specifications. The selected air conditioners should then be installed to the manufacturer's specifications. The reality departs significantly from this scenario. In the end, air conditioners are selected and installed under field conditions that degrade performance.

This study examines three measured factors that affect performance: cooling load, capacity, and attic temperatures. These results were obtained from four intensively monitored, new single-family homes in Phoenix, Arizona. This study found that the most widely accepted sensible heat gain calculation, applied without "safety" factors, overestimated the sensible heat gain for these homes by approximately 50%.

Two of the five air conditioners had sensible capacities significantly below specifications. Both air conditioners with deficient capacity had low airflow and one was seriously undercharged.

Attic temperatures are critical in forced-air distribution efficiency when the ducts are in the attics. On these homes, attic temperatures at peak ranged from 28°F (16°C) to 4°F (2°C) above outside temperatures.

Units: Dual