Abstract Aims The first-generation radiofrequency HotBalloon (RHB) is a size-adjustable single-shot device used in atrial fibrillation. The energy output is determined by its central temperature and not by its balloon surface temperature (BST), thus limiting its efficacy and safety. Therefore, a second-generation RHB was developed to monitor BST and enable BST-controlled ablation. This animal study aims to evaluate the accuracy of a newly developed BST-monitoring system and validate the optimal BST for ablation. Methods and results In Protocol 1, thermocapsules were attached to the superior vena cava (SVC) epicardium. The accuracy of BST monitoring was examined during SVC isolation. In Protocol 2, the efficacy and safety of different BST-controlled ablations were examined. In the acute model, electrophysiological and pathological findings were assessed after energy applications with BST at 51, 54, 57, and 60°C. In the chronic model, the lesion durability and pathological findings were assessed 8 weeks after BST-controlled ablation (57 and 60°C). A significant positive correlation was found between the epicardial temperature and the BST-monitoring value (r = 0.98). In the acute model, all target veins were electrically isolated with BST-controlled ablation at ≥57°C (18/18, 100%). In the chronic model, durable lesions were observed in all veins at 60°C, while 44% of the veins showed reconnection at 57°C. In both pathological analyses, significantly greater lesions were observed at 60°C than at 57°C. There were no significant differences in adverse events between the two groups. Conclusion Balloon surface temperature–controlled ablation at 60°C using the second-generation RHB may be optimal for creating durable lesions without compromising safety.