Introduction Long-term sodium balance studies show that sodium can be temporarily stored and released in tissues, mediated by circaseptan rhythms of aldosterone and cortisol. This complicates the reliability of a single 24-hour urine collection to estimate individual sodium intake. We investigated whether repeated timed urine collection with and without correction for plasma aldosterone is a more accurate alternative for estimating daily sodium intake. Methods We conducted a post-hoc analysis of a metabolic ward study in which 16 healthy male adults consumed a diet with a fixed sodium content (50 or 200 mmol/day) for 7 days. Each day, urine was collected in 4 intervals (7:00-13:00h, 13:00-19:00h, 19:00-23:00h and 23:00-07:00h). Plasma aldosterone was measured at 6:30h, 12:30h and 18:30h. Sodium intakes were estimated by various formulas using 3 timed urines of day 5 to 7. Results During a 200-mmol daily sodium intake, sodium intake estimates based on three repeated timed urine samples and the Toft equation differed 10 [IQR 3-14], 8 [6-19], 36 [16-49] and 20 [10-43] mmol from the actual intake for intervals 7:00-13:00h, 13:00-19:00h, 19:00-23:00h, 23:00-07:00h, respectively. These measurements did not significantly differ from a single 24-hour urine (20 [12-55] mmol). During a 50-mmol daily sodium intake, repeated timed urine collection performed worse than a single 24-hour urine collection. On both diets, correction for plasma aldosterone increased accuracy and sodium intake estimates were significantly more accurate than a single 24-hour urine. Conclusion In a controlled environment, repeated timed urine collection corrected for plasma aldosterone is more accurate than a single 24-hour urine collection.