a total of 13。536 m of piping, and, finally, an internal diam- eter of 4 mm) and a thermodynamic condition (CO2 side) at the inlet of gas cooler of 90 bar and 85 °C。 The air temperature was fixed at 32 ° C and the heat load considered was 600 W。 Ad- ditionally, a case study was run considering two new designs of gas coolers with the focus on CO2 charge reduction and mini- mization in dust accumulation。 The latter was indirect, i。e。 based on a higher air flow velocity and higher fin pitch to reduce blockage by dust in the operating  environments。

3。1。Volumetric flow effect

Simulations were done for each design and considering three air volumetric flow rates。 Table 1 shows the input data speci- fied, which are based on a gas cooler of a cold drink vending machine with 600 W of heat transfer capacity。 Plain fins were considered。

A lateral view of a gas cooler with 2 rows is presented in

Fig。 2 and two points are here highlighted: (i) the counter flow

between the air and CO2 guarantees a high gas cooler perfor- mance and (ii) there is a physical split of the fins between the

where the friction factor is determined using Equations (5) and

(7) above。 For the oil effects on supercritical CO2, the related literature and methods will be presented later in Section   3。5。

two rows to avoid conduction from the hot zones to the cold zones which can otherwise adversely influence the gas cool- er’s  heat  transfer performance。

Fig。 1 – Flowchart characterizing the inputs, the loops of iteration and the convergence parameters implemented in the simulation code。

Fig。 3 shows the CO2 side local heat transfer per unit of tube length, i。e。 the heat transfer determined  for  each increment of length (simulated control volume), as a function of the air volumetric flow rate。 As expected, a higher local heat trans- fer is observed when the volumetric flow is increased and, consequently, a lower tubing length was necessary (6。88, 6。40 and 6。05 m, respectively)。 However, and more important, it is worth mentioning that only two rows are more than enough to handle the desired heat transfer rate, which means the current base case gas cooler is   oversized。

Fig。 4a and b shows respectively the CO2 side heat transfer coefficient as a function of the piping position and the CO2 tem-

perature。 It is quite evident the very strong influence of the supercritical CO2 thermophysical properties on the heat trans- fer coefficient when the fluid’s temperature is in the vicinity of the pseudo-critical point (temperature at which the CO2 spe- cific heat has its maximum value)。 A sharp peak value of the heat transfer coefficient highlights the importance of using a reliable and well thought out heat transfer correlation that ac- counts for these sharp changes in the physical properties (and updating these local properties within the simulation code for the local wall and fluid  temperatures)。

Concerning the air side heat transfer coefficient, the fol- lowing values were predicted when increasing the volumetric

Table 1 – Input data for the evaluation of volumetric flow rate effect on the gas cooler overall performance。

Current Geometry Inlet air and working fluid conditions

Inner diameter [m] 0。004 Working fluid

Outer diameter[m] 0。005 Pressure [bar] 90

Tube spacing in flow direction [m] 0。01819 Temperature [°C] 85

Tube spacing transverse to flow direction [m]