This paper investigates the potential of providing intra-hour load balancing services using aggregated heating, ventilating, and air-conditioning (HVAC) loads. A directload control algorithm is presented. A temperature-priority-list method is used to dispatch the HVAC loads optimally to maintain customer-desired indoor temperatures and load diversity. Realistic intra-hour load balancing signals are used to evaluate the operational characteristics of the HVAC load under different outdoor temperature profiles and different indoor temperature settings. The number of HVAC units needed is also investigated. Modeling results suggest that the number of HVAC units needed to provide a ±1-MW load balancing service 24 hours a day varies significantly with baseline settings, high and low temperature settings, and outdoor temperatures. The results demonstrate that the intra-hour load balancing service provided by HVAC loads meets the performance requirements and can become a major source of revenue for load-serving entities where the two-way communication smart grid infrastructure enables direct load control over the HVAC loads.