We investigated habituation effects during thermal quantitative sensory testing (tQST) using 8 repetitive measurements for thermal detection and pain thresholds. The same measurements were repeated two days later. 39 healthy subjects and 36 patients with chronic non-neuropathic pain syndromes (migraine, tension-type headache, non-radicular back pain) were enrolled. The pain intensity was assessed using an 11-point (0-10) numerical rating scale. Measurements correlated significantly over the two days in both groups (r=0.41...0.62). Warm detection (WDT) and heat pain threshold (HPT) revealed no significant differences over these days. Cold detection (CDT) and pain thresholds (CPT) showed significant differences but these were small compared to the range of normal variability (CDTDelta -0.28 degrees C; CPTDelta 1.51 degrees C). On both days, WDT showed no habituation during measurements. Although there was a small difference in CDT and CPT between first and second measurement, there was no habituation beyond the second stimuli. In contrast, HPT significantly increased between first and sixth stimuli, indicating pronounced habituation. Average HPT of first to third measurement was significantly lower than HPT of the fourth to sixth assessment (45.9 degrees C; 47.7 degrees C) with a good day-to-day repeatability. Repeatability and habituation was identical in both groups. Ongoing pain intensity in the patient groups correlated significantly with CDT/WDT but not with CPT, HPT, indicating that ongoing pain might suppress the sensitivity to non-painful stimuli. In summary, tQST proved a reliable diagnostic tool for clinical practice. Day-to-day differences were small but without clinical relevance. Habituation was most pronounced for HPT, probably due to peripheral fatigue of the receptors.