This paper identifies important faults and their performance impacts f
or rooftop air conditioners. The frequencies of occurrence and the rel
ative costs of service for different faults were estimated through ana
lysis of service records. Several of the important and difficult to di
agnose refrigeration cycle faults were simulated in the laboratory. Al
so, the impacts an several performance indices were quantified through
transient testing for a range of conditions and fault levels. The tra
nsient test results indicated that fault detection and diagnostics cou
ld be performed using methods that incorporate steady-state assumption
s and models. Furthermore, the fault testing led to a set of generic r
ules for the impacts of faults on measurements that could be used for
fault diagnoses. The average impacts of the faults on cooling capacity
and coefficient of performance (COP) were also evaluated. Based upon
the results, all of the faults are significant at the levels introduce
d, and should be detected and diagnosed by an FDD system. The data set
obtained during this work was very comprehensive, and was used to des
ign and evaluate the performance of an FDD method (Breuker 1997a) that
will be reported in a future paper.