A research program at Brookhaven National Laboratory (BNL) has studied ground coupling, i.e., the use of the earth as a heat source/sink or storage medium for solar-assisted and stand-alone heat pump systems. As part of this research program, five serpentine earth coil experiments were operated between December 1978 and September 1981. Heat was added to or removed from the earth coils according to weekly schedules based on computer simulations of solar-assisted and stand-alone, ground-coupled heat pump systems operated in the local (New York) climate. Each earth coil was operated according to a different control strategy. This paper presents experimental results from these experiments for the period December 1978 to April 1981, and compares these results to those generated by a comptuer model, GROCS, developed at BNL. The model is found to provide a reasonably good fit to the data, for the most part, using the experimental undisturbed soil thermal properties. In some cases, the use of a lower soil thermal conductivity provides a better fit, particularly during summer months when heat was added to the ground. Thus, given soil properties, GROCS can be used to predict earth coil performance. If given earth coil performance, the model can predict soil thermal properties. Serpentine earth coils are found to be suitable to provide auxiliary heat or heat rejection for solar heat pump systems. In fact, earth coil-based, stand-alone, ground-coupled heat pump systems can provide all heat needed for winter space heating and all heat rejection required for summer space cooling with no need for any auxiliary heating. Subfreezing winter operation is necessary for shallow earth coils in cold climates. No deleterious effects to the ground were observed from the long-term operation of these experiments.
