The simulated TC intensity and structure were strongly sensitive to colder tropopause temperatures using only longwave radiation, but were less sensitive using full-radiation due to the reduction in net radiative tendencies from shortwave radiation. Colder tropopause temperatures resulted in deeper convection and more intense storms on average, but the maximum intensity was sensitive to small boundary layer moisture perturbations in the initial conditions. The deeper convection led to increased local longwave cooling rates but reduced outgoing longwave radiation at cloud top, such that from a Carnot engine perspective, the radiative heat sink is actually reduced in the stronger storms. We hypothesize that a balanced response in the secondary circulation described by the Eliassen equation arises from the upper troposphere cooling/heating anomalies that leads to stronger eyewall updrafts with enhanced upper level vertical mass flux, stronger outflow, increased ice species, and taller convection. Warmer, lower tropopause heights elicits the reversed effect and weaker storms. Shortwave radiative heating tends to damp this effect by reducing the overall heating and cooling anomalies. These results suggest that the intensity response to a colder tropopause is consistent with PI theory, but that the physical mechanism through which this response occurs is more complex than the simple concept of an outflow heat sink in a Carnot engine. The results of this study further suggest that cloud-radiative feedbacks are important not just in long-term simulations of TCs in radiative-convective equilibrium, but may have a non-negligible impact on weather timescales.