Abstract
A study of energy retrofit by combined heat pipe heat exchanger with cold condensate was carried out and results are presented in this paper. The parameters of study including inlet water temperature, water flow rate, primary air velocity, and exhaust air velocity on systems performance are experimentally investigated. Experiment results indicate that for a small condensate flow rate, both the latent heat of exhaust air and sensible heat of condensate can simultaneously be utilized for cooling and dehumidifying primary air, which is expected to be efficient for energy retrofit in central air conditioning systems.
Introduction
In air conditioning systems, there is a minimum fresh outdoor air requirement for conditioned spaces in order to maintain an appropriate indoor thermal comfort level and a healthy environment. In the meanwhile, a great deal of indoor air must be exhausted outdoors, resulting in energy loss if the heat or coolness of exhaust air is not retrofitted. Especially, in tropics and subtropics with hot and humid climates, indoor moisture load mainly comes from outdoor ventilation air. 1 Energy saving and alleviating loads of cooling coils or other refrigeration system, the cooler and drier exhaust air should be utilized to precool outdoor fresh air. Different types of heat exchangers have been widely used for coolness recovery in air conditioning systems, such as heat pipe heat exchangers (HPHE), heat recovery wheels, runaround coils, and plate-to-plate heat exchangers.2–6
A heat pipe is a very high efficiency heat exchanger and requires no external driving force other than a small temperature difference between its two ends of evaporation and condensation. Since 1970, a series of research work on energy retrofit by using HPHE have been carried out.7,8 The research work on HPHE for energy savings are mainly in two different ways including investigation of energy savings and dehumidification enhancement and investigation of incorporating HPHE into evaporative cooling systems.7–17
In conventional air conditioning systems, the air entering cooling coils is overcooled for satisfying latent load and consequently extra heat must be supplied to reheat the overcooled air before entering the space in order to satisfy the requirement of temperature and humidity settings. An HPHE can be installed in an air conditioning system, wherein the evaporator section of the HPHE locates before the cooling coils and the condenser section of the HPHE locates after the cooling coils. Therefore, before passing the cooling coils, primary air can be precooled by the evaporator side of the HPHE, and the heat gained in the evaporator end of HPHE can be transferred to the condenser end of HPHE, where the overcooled air bypassing cooling coils can be reheated. By precooling the primary air to a temperature being more close to its dew point temperature, the moisture removal capability of cooling coils should be improved. Besides, additional heat for reheating the overcooled can be decreased or even eliminated.7–17 Furthermore, the effects of HPHE on space temperature and humidity control, power consumptions, and peak demand for HVAC systems in residential and industrial buildings are investigated. It is suggested that applications of HPHE can improve energy savings and economic benefits despite of some technical problems, e.g. pressure drop across HPHE.
Evaporative cooling is a simple but effective way of heat transfer, which has been widely used in direct and indirect evaporative coolers.18–28 For direct evaporative cooler,18–23 water is directly in contact with dry air, and there are sensible and latent heat transfer between water and dry air by the virtue of temperature difference and humidity difference. For indirect evaporative cooler,18,24–27 the coolant will not be contaminated and can be kept at a constant humidity, as the coolant is separated from dry air and water. In air conditioning systems with dry exhaust air, HPHE can also be combined with evaporative cooling, while water is sprayed onto the surface of condenser end of HPHE, with exhaust air and primary air passing through the condenser end and evaporator end of the HPHE, respectively.29,30 On the condenser end of HPHE, there are sensible and latent heat transfer between exhaust air and water, and the coolness is transferred to the evaporator end, where primary air is precooled. The study showed that utilization of HPHE incorporated into evaporative cooling system has several advantages, including flexible adaptability to packaged air conditioning systems and large flow rate built-up system, system redundancy, improved indoor air quality and energy efficiency, and electric demand reduction for building cooling load. 29 In dry climatic zones, it is suggested that HPHE systems combined with indirect evaporative cooling system can relieve the load of mechanical cooling systems.
However, literature review indicates that only limited research work on utilization of HPHE for retrofitting energy mentioned above has been carried out in subtropical or tropical climates. 7 In these hot and humid climates such as Hong Kong and Singapore, a cooling coil would produce a great amount of condensate with temperature around 15℃, which can provide latent cooling as well as sensible cooling capacity. Therefore, condensate can be collected together and supplied as water source with adjustable temperatures for an indirect evaporative cooling system in conjunction with HPHE. In this paper, an experimental study of indirect evaporative cooling combined with an HPHE has been carried out.
Experimental setup
A schematic diagram of the test apparatus is shown in Figure 1. Water flowing through a chiller machine was drained into the condenser section of a HPHE, where the cool and dry exhaust air from a conditioned space passes. In the meanwhile, hot and wet primary air flows bypass the evaporator section of the HPHE. Water draining into the surfaces of condenser section evaporates into the exhaust air steam and provides both sensible and latent cooling capacity. Its cooling capacity absorbed in condenser section is released to the evaporator section for cooling and dehumidifying the primary air. In this way, both the energy efficiency and humidity control of the air conditioning system are improved.
Schematic diagram of an indirect evaporative system combined with an HPHE.
The design parameters in this experiment are determined with due consideration. Air from conditioned room is adopted as exhaust air directly, and its velocities passing the condenser section of HPHE should be less than 2 m/s in order to prevent water flooding. It can happen at high velocities.31,32 Primary air is adopted as the outdoor design conditions in Hong Kong at dry bulb temperature of 33℃ and relative humidity (RH) 68%. 1
The dry bulb temperatures and wet bulb temperatures of exhaust air and primary air are measured with eight temperature sensors pt100, and the inlet and outlet water temperatures are measured with two precalibrated thermocouples. All the measured data are logged into a computer by an Agilent Data Logger System. The average velocities of exhaust air and primary air are recorded with a Pitot-Tube and the flow rates of water are measured by a Dataflow Compact Transmitter.
Heat balance equations of the experiment
The heat transfer process in the experiment is complicated with mass transfer. On the condenser side of the HPHE, there are sensible and latent heat transfer between the exhaust air and water films on the surface of HPHE, while on the evaporative side of HPHE, there are heat and mass transfer between primary air and condensate out of primary air enclosing on the surface of evaporative section of HPHE (Figure 2).
The heat transfer among exhaust air, water, and primary via HPHE.
In Figure 2, the sensible heat transfer QS between exhaust air and water film is
The latent heat transfer QL between exhaust air and water film is
Like the heat transfer on the condenser side of the HPHE, if the condensate coming out from primary air encloses the surface of evaporative section of the HPHE, the heat transfer between water and primary air through HPHE is expressed as
In addition, the heat transfer between water film and condensate via HPHE can be written as follows
The energy balance of spraying water exchanging heat with exhaust air and primary air is
Experimental results
In the experiments, the variations of primary air dry bulb temperature and specific enthalpy at outlet and inlet are investigated with different performance parameters, including inlet water temperatures, water flow rates, exhaust air velocities, and primary air velocities. Dehumidification rates of primary air G are calculated as
Test with varying inlet water temperature
Variations of inlet water temperature on the system performance are investigated. The inlet water temperature varies from room temperature to the temperature of condensate, around 15℃. In Figures 3–6, the heat recovery performance with varying inlet water temperature is shown. The exhaust air velocity is fixed at Ve = 1.09 m/s and inlet exhaust air conditions are tei = 22.5 ± 0.1℃, RH = 57 ± 1%. The primary air velocity is fixed at VP = 1.01 m/s and at a dry bulb temperature tPi = 33 ± 0.1℃, RH = 68.5 ± 1%. The water flow rates are kept constant at 0.04 kg/s. In Figure 6, the instrument for temperature measurement was a thermocouple connected to a transducer. The transducer had oscillations in the range 18–21℃. The curve exhibited a little bit odd. Similar phenomenon is observed in Figure 9.
Temperature differences with different inlet water temperatures. Enthalpy differences with different inlet water temperatures. Ratio with different inlet water temperatures. Dehumidification rate variations with different inlet water temperatures. Temperature differences with varying water flow rates. Enthalpy differences with varying water flow rates. Dehumidification rate variations with different water flow rates.






For the primary air that is cooled and dehumidified, its temperature difference and enthalpy difference (ΔH) at outlet and inlet both increase with lower inlet water temperature. For the exhaust air that is also cooled down, its specific enthalpy still increases at outlet due to the latent heat transfer between exhaust air and water. Therefore, the latent heat of water evaporation is partly used to cool down exhaust air itself. This phenomenon is much more obvious since its specific enthalpy difference becomes smaller with lower inlet water temperature (Figures 3 and 4).
At low inlet water temperatures, sensible heat capacity of water is the main contribution for cooling and dehumidification of primary air. On the other hand, the inlet water could also be cooled down while its temperature is higher than a critical temperature
R < 1: No contribution of water sensible heat. From Figure 5, it can be seen that the effect of sensible heat from water on cooling and dehumidifying primary air is strengthened when inlet water temperature decreases. Furthermore, the dehumidification rate is also promoted with cooler inlet water temperatures, i.e., more condensate from the primary air (Figure 6).
As mentioned previously, when inlet water temperature gets cooler, the system performance on cooling and dehumidifying primary air could be improved with the increasing inlet water sensible heat capacity, although total heat transfer between exhaust air and water film decreases due to a smaller enthalpy difference. Therefore, it would be very useful to continue the experiment with condensate, an inlet water temperature at about 15℃, while there is enough condensate collected for heat recovery.
Tests with condensate
Test with varying water flow rate
Experimental results with different water flow rates are shown in Figures 7–9, with inlet exhaust air at tei = 22.1 ± 0.1℃, RH = 60 ± 1%, Ve = 0.9 m/s, and inlet primary air at tPi = 33 ± 0.1℃, RH = 68 ± 1%. The inlet water temperature is set at twi = 14.8 ± 0.1℃.
The temperature difference and enthalpy difference of primary air both increase while more water is drained onto the surface of the condenser section of HPHE. This is because increasing water flow rate can improve the heat transfer coefficient between the water and primary air via HPHE. Furthermore, the outlet water temperature could be lowered with increasing water flow rate. As a result, the average temperature difference between the condenser section surface and evaporative section surface of HPHE increases, which can improve the cooling and dehumidification of primary air. From Figure 9, it can be found that the dehumidification rate is also improved with a higher water flow rate. The enthalpy difference of primary air is also increased because of increasing sensible heat transfer as well as dehumidification (Figure 8).
However, the enthalpy difference of exhaust air decreases with increasing water flow rate, which means that the latent heat capacity of exhaust air on cooling and dehumidifying is reduced. It can be explained as the average water temperature is lowered with a higher water flow rate. The heat transfer due to enthalpy difference between the saturated air around water film and exhaust air is weakened. Specifically, at a high water flow rate, there is heat transfer from exhaust air to water film with a negative enthalpy difference for exhaust air.
Test with different primary air velocities
The experimental results with different primary air velocities are shown in Figures 10–12, with inlet exhaust air at tei = 22.5 ± 0.1℃, RH = 56 ± 1%, Ve = 1.09 m/s, inlet primary air at tpi = 33 ± 0.1℃, RH = 67 ± 1%, and inlet water at twi = 14.8 ± 0.1℃, Temperature differences with different primary air velocities. Enthalpy differences with different primary air velocities. Dehumidification rates with different primary air velocities.


It is shown that the temperature difference of primary air decreases as more primary air is introduced. Its enthalpy difference and dehumidification rate both increase with more primary air (Figures 10 and 12). In addition, it can also be found that the total heat transfer between water and exhaust air is small at a low primary velocity as the enthalpy difference is small (Figure 11). It is suggested that the exhaust air makes a little contribution to cooling and dehumidifying primary air. The temperature difference of water would increase with increasing primary air in illustrated in Figure 10. Subsequently, the enthalpy difference of exhaust air can be increased due to increasing enthalpy difference between water film and exhaust air (Figure 11).
Test with different exhaust air velocities
Figures 13–16 show the impacts of exhaust air velocities on system performance with inlet exhaust air at tei = 22.6 ± 0.1℃, RH = 54 ± 1%, Vp = 0.94 m/s, inlet primary air at tpi = 33 ± 0.1℃, RH = 67 ± 1%, and inlet water at twi = 14.8 ± 0.1℃, Temperature differences with different exhaust air velocities. Enthalpy differences with different exhaust air velocities. Dehumidification rates with different exhaust air velocities. Sensible and latent heat transfer for exhaust air at different velocities.



Experimental results indicate that the variations of exhaust air velocities have little impact on cooling and dehumidifying primary air. In Figures 13 and 14, the temperature and enthalpy difference of primary air both slightly increase while more exhaust air is introduced. In Figure 15, variations of dehumidification rate are also not obvious with increasing exhaust air velocity.
With increased exhaust air intake, the latent heat transfer between exhaust air and water is strengthened. However, sensible heat transfer of exhaust air is also increased, i.e., most of the increased latent heat capacity is to cool the exhaust air itself (Figure 16). As a result, the variation of the total heat transfer between exhaust air and water is very small, and the enthalpy difference of exhaust air is only slightly changed with different exhaust air velocities. Therefore, the variations of exhaust air flow rate have little effect on system performance.
Test without exhaust air
As investigated in Test with different exhaust air velocities section, the effects of exhaust air velocities are not obvious. Specifically, experiments without exhaust air bypassing the condenser section of the HPHE are carried out and compared against the experimental results in Test with varying water flow rate section. Results are presented in Figures 17 and 18.
Enthalpy differences with varying water flow rates at Ve = 0 and Ve = 0.9 m/s. Dehumidification rates with varying water flow rates at Ve = 0 and Ve = 0.9 m/s.

It is shown that the system behavior is different with increasing water flow rate at
However, the effects of exhaust air are weakened with increasing water flow rate. Especially, at higher water flow rates, there is heat transfer from exhaust air to water with a negative enthalpy difference of exhaust air as mentioned in Test with varying water flow rate section. The temperature difference, enthalpy difference, and dehumidification rate at
Because of limited water condensate collected from most A/C systems, it would be better to employ water and exhaust air simultaneously for cooling and dehumidifying primary air, as both of the latent heat capacity of exhaust air and the sensible heat capacity of water can be reutilized.
Conclusions
In this paper, an HPHE combined with water and exhaust air is used for heat recovery in an indirect evaporative device. By distributing water onto the condenser section of the HPHE, the latent heat capacity of exhaust air and the sensible heat capacity of water can both be retrofitted to cool and dehumidify primary air.
Experiments with different inlet water temperatures indicate that the lower the inlet water temperature is, the better the system behaves. However, at low inlet water temperatures, the ratio of the latent heat capacity of exhaust air employed to cool and dehumidifying primary air is small, because the heat transfer between the water film and exhaust air is not considerable due to small enthalpy difference.
Subsequently, the inlet water temperature is set at about 15℃, which is the usual temperature of condensate from most A/C systems. Experiments with varying water flow rate, primary air velocity, and exhaust air velocity on system performance are investigated. There are several major points concluded from this study.
The effect of sensible cooling and dehumidifying of primary air is reducing (R < 1) with inlet water temperature higher than 15℃ (Figure 5); The dehumidification rate of primary air deduced from Figure 6 is about 0.24 g/min per 1℃ increasing of inlet water temperature (r2 = 0.957); Water flow rate is correlated positively in logarithmic form with dehumidification rate (g/min = 1.7322 × ln (kg/s) + 9.977) and saturated at 0.055 kg/s. The correlation coefficient (r2) is 0.9488 (Figure 9); Primary air velocities are correlated positively in logarithmic form with dehumidification rate (g/min = 1.4162 × ln (m/s) + 3.9298) and saturated at 1.8 m/s. The correlation coefficient (r2) is 0.9839 (Figure 12); The exhaust air velocities are not correlated positively with differences of enthalpy (Figure 13−15); The influences of exhaust air on cooling and dehumidifying primary air are not obvious, because the increased latent heat transfer of exhaust air with more exhaust air could not be transferred to cool and dehumidify the primary air via the HPHE; instead of the exhaust air itself is further cooled down with increasing exhaust air velocity; and It is suggested that the exhaust air and condensate should be combined together in applications as the amount of condensate is not remarkable in most air conditioning systems.
Nevertheless, problems also exist in this study. The latent heat capacity of exhaust air could not be effectively used to cool and dehumidify primary air through the HPHE in the experiments, such as the increasing latent heat transfer with more exhaust air; the water distribution onto the condenser section surface of HPHE is not very well and only part of its surfaces are enclosed with water film, which may also have negative impacts on the system performance.
Footnotes
Conflict of interests
None declared.
Funding
The author acknowledges the funding supports G-YL28 and G-YBA8 from The Hong Kong Polytechnic University for the work reported in this paper.
