| A procedure is developed for obtaining the uncertainty, which properly accounts for correlations between input parameters, in the computed performance characteristics (rating, pressure drops, and fluid stream exit temperatures) of a cross-flow heat exchanger. The method is applied to a typical cross-flow heat exchanger analysis problem and results are compared to those of earlier studies which neglected correlations between input parameters. To encourage the use of uncertainty analysis in heat exchanger analysis and design, guidelines are given for finding the available uncertainty information or estimating uncertainties when published uncertainty information is not available. Results indicate that if an uncertainty analysis is to be done, the precalculated values of the geometric parameters hydraulic diameter, surface area density, and fin-area-to-total-area ratio normally obtained from the heat exchanger surface database should not be used as input parameters in the analysis, but rather should be calculated from the fundamental geometric parameters such as parting plate spacing and fin pitch.; Additionally, a computer program is developed which accounts for the combined effects of flow nonuniformity, lateral/longitudinal wall conduction, and stacking effects in predicting the thermal performance and pressure drops for a typical cross-flow heat exchanger. Flow models are developed for modeling typical nonuniform flow configurations (randomly distributed with specified standard deviation, fully-developed 1/7 power law, and flow downstream of fans, elbows and short transitions). Results are compared with an "ideal" heat exchanger to determine the loss in thermal performance and increase in pressure drops due to these nonideal conditions. Results show that the stacking effect is strongly coupled with the flow nonuniformity and conduction effects and that some nonuniform flow configurations can severely intensify the stacking effects.; Uncertainties arising from neglecting these departures from ideal conditions are always in the "loss" direction and are said to be "one-sided" or asymmetric. These asymmetric uncertainties are successfully incorporated into the cross-flow heat exchanger uncertainty analysis by adapting a recently developed technique (to replace the old standard in the ANSI/ASME PTC19.1-1985) for determining the uncertainty in an experimental result when some of the measured variables have asymmetric systematic (or bias) uncertainties. |
