Home | Business News | Browse by Publication | H | HVAC & R Research

The effect of membrane deflections on flow rate in crossflow air-to-air exchangers.

Article Excerpt
INTRODUCTION

One way to increase the energy use efficiency of typical commercial buildings is to recover waste energy from exhaust air and use it to condition the supply ventilation air. Since air temperature and relative humidity are controlled for the best comfort conditions in buildings, recovery of both the thermal or sensible energy and moisture or latent energy is important. Energy and heat recovery systems using energy wheels, crossflow heat exchanger plates, and heat pipes have been widely used to lower heating and cooling costs for buildings (Besant and Simonson 2003; Simonson 2007). Recently, new devices have been introduced in the HVAC industry that employ semipermeable membranes to transfer both water vapor and heat (Fan et al. 2006). A numerical investigation of this type of crossflow air-to-air exchanger using semipermeable membranes has been presented by Zhang and Jiang (1999). These exchangers have designs similar to traditional plate exchangers except that the nearly rigid plates are replaced by flexible semipermeable membranes that need more mechanical support. A schematic of the airflow streams of a typical membrane, crossflow, air-to-air exchanger is shown in Figure 1.

[FIGURE 1 OMITTED]

The membrane material is typically a hydrophilic desiccant-coated paper or plastic that is permeable to water vapor--but not air--so it facilitates the transfer of latent energy by water diffusion as well as sensible energy by heat conduction between the supply and exhaust streams of an HVAC system. The transfer of both latent and sensible energy using a crossflow membrane exchanger has been shown by Niu and Zhang (2002) to be significantly more cost-effective than a crossflow heat exchanger that is used for sensible energy transfer only. This increase in total energy performance effectiveness, compared to typical sensible energy effectiveness for heat exchangers, makes semipermeable membrane air-to-air exchangers that transfer heat and moisture a better design option to meet the increasing demands for efficient energy use in buildings.

For a given airflow rate, the sensible and latent effectiveness of a crossflow plate exchanger will increase as the heat and moisture exchange surface areas increase. Thus, typical exchanger designs use thin flow channel spacing to maximize the number of flow channels per unit volume of exchanger and, therefore, surface exchange area per unit volume. While narrowly spaced flow channels increase the surface area per unit volume, the airflow becomes laminar, and there are manufacturing and operating difficulties in achieving and maintaining uniform channel sizes due to manufacturing tolerances and operating condition changes. The effects of manufacturing tolerances have been investigated by several authors and are reviewed by Shang and Besant (2005).

Shang and Besant (2005) show that for rigid parallel plate regenerative wheel exchangers with fully developed laminar flow at a specified flow rate, the ratio of pressure drop between two channels varies as the cube of the ratio of their hydraulic diameters. Conversely, for a specified pressure drop, the ratio of the flow rate through each channel will vary with the cube of the hydraulic diameter ratio. If the flow channels can be considered to be symmetric cylinders, the cubic relationship is replaced by a fourth-power relationship. Shang and Besant (2005) used these cubic or fourth-power relationships to investigate the effect of spacing variations due to random variations in channel sizes to get the change in overall exchanger pressure drop compared to an exchanger with no variations but the same airflow rate. They found that both the overall exchanger pressure drop and the effectiveness decreased with increasing flow channel size variations.

An important question that needs to be considered in this research for membrane plate exchangers is how will flow channel spacing variations due to manufacturing tolerances or membrane deflections alter their performance? In this research we consider only the effect of average membrane deflections on flow rate for a given pressure drop or, conversely, pressure drop for a given flow rate. In addition to membrane deflections caused by a differential air pressure across the membrane, any pre-slack or pre-stress in the membrane during manufacturing of the exchanger will affect the membrane deflection. The effects of membrane deflection due to pressure differential across the membrane and pre-slack or pre-stress are superimposed or added to give a total deflection.

The objective of this paper is to show the effects of membrane deflection on the air mass flow rate or pressure drop across the air-to-air exchanger. Membrane deflections cause the average hydraulic diameter to change in each flow channel, and this change will cause the flow rate to change for a fixed pressure drop across the exchanger or, conversely, the pressure drop to change for a given total flow rate through the exchanger. The effect of pre-tensile stress or initial slack of the membrane on the exchanger pressure drop is also investigated. Airflow through the exchanger is assumed to be...

View this article FREE - Now for a Limited Time, try Goliath Business News
Free for 3 Days!