The aim of the study was to develop and validate a feline urinary collection system for accurate 24 h urine output measurement and glomerular filtration rate (GFR) evaluation. We hypothesized that precise identification of urination time improves the accuracy of the collection system. In a group of nine cats, urinary volume and micturition times were repeatedly recorded for up to 48 h using purpose-built collection trays containing a temperature data logger. Collection time was determined both with and without using the data loggers on 22 occasions and agreement between estimated 24 h urine outputs obtained with the two calculation methods was evaluated. GFR was repeatedly measured by endogenous urinary creatinine clearance on three occasions. Twenty-four-hour urine output was measured in 98.5% of the attempted collections (300 cat-sampling days). Sensitivity and specificity of the detection system were 97.8% and 100%, respectively. Mean 24 h urine output was 12.4 ± 4.94 mL/kg/day and mean intra-cat between-days coefficient of variation (CV) was 16.6 ± 5.6% when data loggers were used. The absolute relative volume error between the two calculation methods ranged from 0% to 131%. Median absolute relative [interquartile range] error was 9.1% [3.25-19.8]. Bias was -1.3% and lower and upper limits of agreement were -39.7% and 35.2%, respectively. Mean estimated GFR was lower than previously reported with comparable urinary clearance methods (1.92 ± 0.37 mL/min/kg) and mean within-cat CV was 12 ± 6.9%. The system was simple in design, readily affordable, allowed normal micturition behaviour and reduced intra-animal variability in 24 h feline urine collection.