Abstract: The study shows the experimental and theoretical analysis of a fast beverage chiller based on the principle of a thermoelectric refrigeration and it was discovered theoretically that for a cooling time of 2 min and a beverage size of 474 mL (16 Oz), 6 TEC modules and hence 12 heat sinks (one for each side of the TEC module) were needed. However by increasing cooling time to 4 min and decreasing beverage size to 325 mL, 4 TEC modules and hence 8 heat sinks could be used. Comparison were also made between the beverage chiller’s cooling time with cooling times obtained from the freezer space and cold space of a household refrigerator. All three tests were carried out on 325 mL of water in a glass jar. The result shows that for the refrigerator freezer space, the temperature of the water decreased linearly with increasing time. However for the beverage chiller, the water temperature decreased exponentially with increasing time.

INTRODUCTION

A thermoelectric device is one that operates on a circuit that incorporates both thermal and electrical effects to convert heat energy into electrical energy or electrical energy to a temperature gradient. Thermoelectric elements perform the same cooling function as Freon based vapor compression or absorption refrigerators. Energy is taken from a region thereby reducing its temperature. The energy is than rejected to a heat sink region with a higher temperature.

Thermoelectric elements are in a totally solid state while vapor cycle devices have moving mechanical parts that require a working fluid. Thermoelectric modules (Fig. 1) are small sturdy, quiet heat pumps operated by a DC power source. They usually last about 200,000 h in heating mode or about 20 h if left on cooling mode. When power is supplied, the surface where heat energy is absorbed becomes cold; the opposite surface where heat energy is released becomes hot. If the polarity of current flow through the module is reversed, the cold side will become the hot side and vice-versa.

Thermoelectric devices can also be used as refrigerators on the bases of the Peltier effect (Cengal and Michael, 1998). To create a thermoelectric refrigerator (Fig. 1), heat is absorbed from a refrigerated space and than rejected to a warmer environment. The difference between these two quantities is the net electrical work that needs to be supplied. These refrigerators are not overly popular because they have a low coefficient of performance.

Fig. 1:	A thermoelectric refrigerator based on the Peltier effect

Thermoelectric modules can be used as thermocouples for temperature measurement or as generators to supply power to spacecrafts and electrical equipment. Thermo electronic devices are used in a variety of applications. They are used by the military for night vision equipment, electronic equipment cooling, portable refrigerators and inertial guidance systems. These products are useful to the military during war and training because they are reliable, small and quiet. Another advantage to these thermoelectric products is that they can be run on batteries or out of a car lighter. The medical community uses thermoelectric applications for hypothermia blankets, blood analyzers and tissue preparation and storage. The main advantage of thermoelectric devices to the medical community is that the devices allow doctors precise temperature control which is useful in handling tissue samples. Hypothermia blankets are pads that patients rest on during surgery to keep their body at a certain temperature. Thermoelectric devices are probably most well known for their contribution to powering spacecrafts like the Voyager.

Radioisotope thermoelectric generators provided all of the on board electrical power for NASA’s Voyager. The thermoelectric devices proved reliable since they were still performing to specification 14 years after launch. The power system provided the equivalent of 100-300 Watts electrical power and multiples thereof. NASA is now requiring higher efficiency rates out of smaller units. The interaction between thermal and electric phenomena (Seebeck effect (1821), Peltier effect (1834), Joule effect (1841) and Thomson effect (1857)) was known since the 19th century (Rowe and Bhandari, 1983). In 1885 the English physicist J.W Rayleigh outlined the possibility of using thermoelectric devices as electricity generators but his development was totally stopped because of the low efficiency achieved. However, the major advance was made in the 1950s with the introduction of semiconductors as thermoelectric materials. It was observed that they had a high Seebeck coefficient, good electrical conductivity and low thermal conductivity. In those moments thermoelectric refrigeration began to look more promising and Peltier devices were developed for refrigeration applications mostly for the military field. Research on semiconductor thermocouples also led to the construction of thermoelectric generators with a high enough efficiency for special applications.

There was little improvement in thermoelectric materials from the time of the introduction of semiconductor thermo-elements until the end of the twentieth century. However, in recent years, several new ideas for the improvement of materials have been put forward and significant advances are being made (Goldsmid, 2009).

Nowadays, in the civil market, thermoelectric refrigeration has a place in medical applications and scientific mechanisms and devices where accurate temperature control is needed. Nevertheless, there are other applications with great potential in which companies are starting to show interest, e.g., dehumidifiers (Vian et al., 2002), domestic and automobile air conditioning systems, portable iceboxes, domestic refrigerators, devices to transport perishable products, computer processor coolers, etc. For these applications, thermoelectric refrigeration competes with conventional refrigeration systems like vapor compression refrigeration.

For a typical conventional refrigeration system, a temperature difference between the ambient and the cabinet of about 25-30 K at T_h = 300 K is usually required to achieve satisfactory cooling performance. This indicates that the maximum COP of a thermoelectric refrigerator comprised of a commercially available module is around 0.9-1.2. However, the practical COP of a thermoelectric refrigerator is much lower than this because the temperature difference between the hot and cold side of the thermoelectric module is larger than the temperature difference between the ambient and the cabinet. In other words, the hot side temperature is higher than the ambient and the cold side temperature is lower than the cabinet temperature.

For a practical thermoelectric cooling system, the hot side heat exchanger rejects the heat produced on the hot side of the thermoelectric module to the ambient and so reduces the hot side temperature. The cold side heat exchanger removes the heat from the cold region to the cold side of thermoelectric module and so increases the temperature of the cold side. Because the thermoelectric module is very high heat intensity equipment, the high efficiency thermoelectric heat exchangers is necessity.

Use of a heat pipe will not be of benefit for natural convection because the dominant thermal resistance in this case is the convection resistance (Webb, 1998). Water-cooled forced convection heat exchangers have excellent performance. The main drawback of a water-cooled heat exchanger is that it needs a convenient source of cooling water. Without a source of cooling water, a forced convection water heat exchanger would require a pump and radiator and associated fittings and tubing. The added resistance of the radiator would increase the overall resistance.

Air-cooled systems are therefore often more desirable. Many heat exchange systems based on the afore-mentioned forced air convection exchangers and the use of heat pipes have been reported. Sofrata (1996) reported that using a double fan in an appropriate position could significantly increase the efficiency of the forced air exchanger compared to using the single fan in a refrigerator.

A long chimney for a natural-convection heat exchanger may also improve the performance of the refrigerator without the need to use fans that of course require the electrical power input. A novel, air-cooled thermosyphon reboiler-condenser system has been reported (Webb, 1998) and has been used as a heat exchanger of a thermoelectric refrigerator (Gilley and Webb, 1999).

This system is capable of providing very low heat sink resistance values with air cooling and a thermal resistance as low as 0.0194-0.0505 KW^-1 was obtained for cooling a 45 mm² module. The system promises significantly higher COP for thermoelectric coolers than is possible using existing heat exchange technology. Riffat et al. (2001) have reported a thermoelectric refrigeration system which employs a Phase Change Material (PCM) as a cold side heat exchanger for cooling storage and improvement of the COP. The refrigeration system was first fabricated and tested using a conventional heat sink system (bonded fin heat sink system) at the cold heat sink. In order to improve the performance and storage capability, the system was reconstructed and tested using an encapsulated Phase Change Material (PCM) as a cold sink.

Both configurations used heat pipe embedded fins as the heat sink on the hot side. Results of tests on the latter system showed an increased performance. This was because the PCM had a large storage capacity allowing most of the cooling energy to be absorbed by the PCM and therefore, the cold side temperature fell more slowly than when the PCM was not used. During the phase change process, the temperature of the refrigeration system was almost constant until the phase change process was complete. This helped to keep the temperature difference across the thermoelectric module to a minimum, thus improving its performance.

In general, thermoelectric modules are very high heat intensity equipment which need high efficiency heat exchangers to lower the hot side temperature and increase the cold side temperature in order to improve the COP. Use of a greater number of modules would also improve the COP of the system. Use of more modules would reduce the heat load on each module and so lower the heat flux densities of both the hot and cold side of each module.

MATERIALS AND METHODS

Thermoelectric modules parameters selection:

Where:

Fig. 2:	TEC module performance graph (obtained from TEC module spec sheets)

Heat sink sizing: Heat sink design parameters (Fig. 3) given as:

Volumetric flow rate of heat sink fan, v = 300, cfm = 8.50 m³ min^-1 = 0.142 m³ sec^-1. Air velocity between fins:

Where:

Spacing between fins:

Properties of air @ 1 atm. and 28°C, μ = 184.6x10^-7 Ns m^-2

Fig. 3:	Heat sink design parameters

The composite model for forced convection for the plate finheat sink is given by the equation:

where, Nu_i = Ideal Nusselt’s number (equation developed by Teertstra et al., 2001).

Teertstra et al. (2001) modelled the heat sink as (N_fin-1) parallel plate 2 dimensional channels where channel width (space between fins, b) is used as the characteristic length. The Reynold’s number, R_e is thus defined by:

The channel width, length and Reynold’s number are combined to obtain the channel Reynolds number, R_e^*, analogous to the channel or Elenbaas Rayleigh number in natural convection:

Average heat transfer co-efficient,

Fig. 4:	Thermal resistance across heat sink fins

Incropera et al. (2007) where:

Hence:

Heat sink thermal resistance (Fig. 4), R_hs is then given by:

where, η_o = Overall sueface efficiency which characteristic an array of fins and the base to which they are atteched (Incopera et al., 2007).

Where:

where, A_b = Prime surface/base area of spaces between fins:

Hence:

0.049 K/W>>0.05 K/W (Maximum allowable heat sink resistance). Therefore, heat sink design meets requirements.

RESULTS AND DISCUSSION

Experimentation: The following tests were carried out on the fast beverage chiller: Two of the TEC modules were connected to two 12 V computer power supplies and the other two were connected to a variable voltage Lab power supply. The blower was connected to 110 V ac. The readings for the voltage and current are shown in Table 1.

The airflow rate through the two warm compartments were also measured and the readings were shown in Table 2. Thermocouples were installed at various points in the beverage chiller to take the temperatures at these points.

Table 1:	The voltage and current readings

The placement of the thermocouples is shown in Fig. 5. Where, T1 indicates temperature at hot side heat sink base, T_base,hot; T2 measures temperature at cold side heat sink base, T_{base cold}; T3 measures cold compartment temperature, T_cold and T4 measures water temperature, T_water.

Table 3 and 4 shows the experimental results for the beverage chiller when allowed to reach operating temperature with nothing in the cold compartment. The same quantity of water in the same glass jar was then placed in: the freezer compartment of a household refrigerator, the cold space of a household refrigerator. Readings were taken until the desired temperature of 6°C was obtained results shown in Table 5. The following graphs were then plotted to observe the performance of the Beverage chiller.

The design power input obtained in the calculations for the TEC modules used was 98 W. Figure 6 shows the performance of a TEC module variation with input power: Performance in this case refers to the cooling rate Q_c. The Fig. 6 shows that as the input power increases, performance also increases.

Table 2:	The air flow-rate through the two warm compartments

Fig. 5:	The placement of thermocouples (T1, T2, T3 and T4)

Table 3:	Temperature readings for the cold and hot compartment

Table 4:	Temperature readings for the cold and hot compartment when water was tested

Operating at close to the maximum is inefficient; most applications do operate at 40-80% of input maximum power of TEC modules.

For the selected TEC modules 40% input power max is 160.16 W. Therefore, the cooling rate of the beverage chiller would have been improved if the design requirement of 98 W for each TEC module was met. If the available power supplies supported supplying 160 W, this would have increased the cooling rate even more. Inefficient forced convective heat transfer within the cold compartment also adversely affected the cooling rate of the beverage chiller as shown in Fig. 7. A small fan was used to circulate air within the cold compartment. The fan had small cubic feet per minute (cfm) rating and its positioning also was not optimal. Having a fan set-up where the fans circulate air through the fins of all the cold side heat sinks and then directing a blast of cold air over the jar containing the water to be cooled would have resulted in better cooling times.

Figure 8 compares the beverage chiller’s cooling time with cooling times obtained from the freezer space and cold space of a household refrigerator. All three tests were carried out on 325 mL of water in a glass jar. The result shows that for the refrigerator freezer space, the temperature of the water decreased linearly with increasing time. However, for the beverage chiller, the water temperature decreased exponentially with increasing time.

Table 5:	Variation of temperature with time for 325 mL water placed in freezer and cold space of refrigerator

Fig. 6:	Variation of performance with input power for a TEC module

Fig. 7:	Variation of temperature with time of the cold compartment of the beverage chiller and of 325 mL water placed inside the cold compartment

In other words, cooling rate for the refrigerator was constant while for the beverage chiller it decreased exponentially. Figure 8 also shows that the freezer took 61 min to cool the water to 6°C while the beverage chiller took 69 min. It can be seen that for the majority of the cooling time, the beverage chiller was cooling at a faster rate than the freezer.

Fig. 8:	Variation of temperature with time of 325 mL of water placed inside various cold spaces

But by virtue of the exponential cooling versus linear cooling, cooling rate for the beverage chiller was decreasing while the freezer cooling rate was constant throughout the cooling process. This caused the beverage chiller’s cooling rate to eventually reach a point where it was lower than the freezer’s cooling rate. This happened at around 7°C as shown in the Fig. 8 where the lines crossed. It must also be noted that the temperature within the freezer space was measured at -17.4°C while that of the beverage chiller’s cold compartment was on average around 3.9°C (it started at 5.7°C and dropped to 2.7°C during the water cooling process).

Therefore, at the point in time at which the required water temperature of 6°C was attained, the temperature difference between water and cold space was 23.4°C for the freezer and only 3.30°C for the beverage chiller. This shows that although the beverage chiller took more time to cool the water, the heat transfer process from the water to the cold compartment was more efficient than in the freezer. The cold space of the refrigerator was measured at 5.1°C and took over 2 h to cool to 7.2°C which was very much slower than the beverage chiller.

The theoretical analysis showed a supply voltage of 7 V and current of 14 A obtained for the TEC modules from the TEC module performance curves. Which gives a wattage of (7x14) = 98 W. However, to gain this power rating of 98 W, 11.69 v had to be supplied which caused a corresponding current of 8.4 A in the TEC module. Design calculations stipulated a volume flow rate at the beginning of each heat sink of 150 cfm. This design requirement was met. For the left side warm compartment, the flow rate at the beginning of the first heat sink was 530 and 160 cfm at the beginning of the second heat sink. For the right side warm compartment, the flow rate at the beginning of the first heat sink was 880 and 450 cfm at the beginning of the second heat sink. The difference in the flow rates on the left and right side can be explained by the fact that the left side had a longer duct with an extra 45° angle than the right side. Therefore, pressure loss along the duct that supplies the left side would have been greater than that along the duct supplying the right side.

The dimensions and different airflows for heat sink thermal resistance was obtained from the mathematical analysis. The use of excel was instrumental in speeding up this process and the above calculations only represent the final result with the desirable values. Variables that were found to decrease the heat sink’s thermal resistance are as follows: Increasing air flow-rate across heat sinks, Increasing Fin length (H_f), Increasing Fin width (L), Increasing number of fins (n_f), while keeping Heat sink width (W) the same, hence decreasing fin thickness (t_f) also; Increasing fin thickness (t_f), while keeping Heat sink width (W) the same, hence decreasing space between fins (b) also.

CONCLUSION

The theoretical determination of TEC module and heat sinks shows that for a cooling time of 2 min and a beverage size of 474 cc (16 Oz) mL, 6 TEC modules and hence 12 heat sinks (one for each side of the TEC module) were needed. However by increasing cooling time to 4 min and decreasing beverage size to 325 mL, 4 TEC modules and hence 8 heat sinks could be used.