Modeling and Experimental Validation of Area Constrained Ice Storage Systems

J.E. Braun, Ph. D., P.E. Purdue University This paper describes the development and validation of an improved mechanistic model for predicting the thermal performance of an area constrained ice storage system. Experimental data was obtained from a fully instrumented thermal storage test cell located within the Herrick Laboratories, Purdue University. Results for both charging and discharging cycles are presented for a variety of inlet brine temperatures and flow rates. The model predictions agree well with the experimental data. The storage tank heat transfer effectiveness was found to be highly coupled to the flow rate but insensitive to changes in the inlet brine temperature. Supercooling and recalescence were also observed during the charging cycle. NOMENCLATURE variable list A..., = node surface area, ft2 Cp _=specific heat constant pressure, Btu/lb-°F ey'_ =specific heat constant volume, Btullb-°F n:.. = diameter, ft h _ = exterior convection coefficient, Btulhr-ft2_ °F bin =interior convection coefficient, Btufhr-ft2_op L =thermal conductivity, Btu/hr-ft-°F L = node length, ft m_ =mass, lb R_ = Thennal resistance, °F /Btu Rspiral = average spiral radius, ft Ra = Rayleigh Number T= temperature, °F U_ = internal energy, Btu Usl =latent heat of :fusion, Btu/lb INTRODUCTION subscripts, b =brine b,i = brine inlet b,o = brine outlet ice = ice formation m = tube interior 1-w = ice water interface s =storage tube = tube exterior t-w =tube water interface w =water Ice storage systems provide an effective means of reducing peak daytime electrical consumption by shifting operation of the primary cooling equipment to· the nighttime. During a charging cycle, a 25% ethylene glycol and water (brine) solution is pumped through a reciprocating or centrifugal chiller and delivered to the storage tanks at approximately 23°F (-5°C) to 26°F (-3.3°C). During the discharging period; the chillers, ice storage tanks, or a combination of both provide chilled brine to the building at a temperature between 38 and 45°F (3.3-7.2°C). The model described in this paper will predict the performance of area constrained ice storage tanks. The tank used to validate the model consists of an insulated cylindrical enclosure surrounding rougWy 15,000 lineal feet (4572 m) of 112 inch (1.3 em) o.d. polyethylene tubing. This tubing is divided into approximately 65 parallel circuits. The tubes are secured to plastic spacers which provide roughly 1/2 inch (1.3 em) clearance between adjacent circuits. Each individual circuit forms a spiral located within a horizontal plane. Adjacent circuits are connected in an alternating sequence between two separate supply and return manifolds. This counterflow arrangement promotes uniform ice growth throughout the storage tank. The remaining enclosure volume is filled with 830 gallons (3.14 m3) of water. A brine solution is pumped through the inside of the tubes and ice is either formed or melted outside the tubes. The close tube spacing minimizes the ice thickness and its corresponding thermal resistance. Near the end of the cycle, the growing ice formations intersect causing a rapid loss in surface