Page 109 - Defrosting for Air Source Heat Pump

Modeling study on uneven defrosting 101

4.3.1.4 Case 1

Fig. 8 shows the measured tube surface temperature at the exit of each circuit during
defrosting in the previous experimental study in Case 1 in Section 3.3. It could be seen
that the temperature order of the three circuits was kept at T 1 > T 2 > T 3 during
defrosting. This is because of the negative effects of the downward flowing melted frost.
Experimental results show that the defrosting durations for the three circuits from top to
bottom were 172, 182, and 186 s, respectively. In other words, the defrosting durations
for the top and the middle circuits were 92.5% and 97.8%, respectively, of that for the
bottom circuit. To alleviate the uneven defrosting, Study Case 1 was then designed,
where the modulating valve for the bottom circuit was fully opened and the openings
of the modulating valve for the top and middle circuits were set at 92.5% and 97.8% of
full opening, respectively. In this way, the heat supplies to the three circuits via the sup-
ply of refrigerant during defrosting were no longer the same, and the assumed refrigerant
mass flow rates to each circuit during defrosting are 95.6%, 101.1%, and 103.3% of their
previous values, respectively, as shown in Fig. 4.20.
Fig. 4.22 shows the assumed refrigerant mass flow rate in the three circuits during
defrosting in Study Case 1. The trends of refrigerant mass flow rates in the three cir-
cuits during defrosting at three stages are the same as those shown in Fig. 4.13. How-
ever, their peak values at 160 s into defrosting were 10.10 g/s for Circuit 1, 10.62 g/s
for Circuit 2, and 10.83 g/s for Circuit 3, respectively. During defrosting, the mass
flow rate order was always at R 3 > R 2 > R 1 , which met the designed experiment con-
ditions in Table 4.2 and the changes in the proportion of the refrigerant distribution
shown in Fig. 4.20A.

4.3.1.5 Case 2

As shown in Fig. 4.21, the results from the previous experimental study also revealed
that the defrosting duration for Circuit 1 was the shortest. Hence, it was also possible
to vary the heat input to the three refrigerant circuits by fully closing the modulating
valve on Circuit 1 when its tube surface temperature at exit reached 24°C, which
means its defrosting process was terminated. Therefore, in Study Case 2, it was
designed that the three valves on the three circuits were fully open at the start of
defrosting. When the defrosting of Circuit 1 was terminated, the modulating valve
on it would be fully closed. Consequently, more refrigerant would flow into the other
two refrigerant circuits to speed up their defrosting. Fig. 4.20B illustrates the changes
in the proportion of the refrigerant distribution into each circuit, and Fig. 4.23 shows
the assumed refrigerant mass flow rates to each circuit during defrosting in Study Case
2. It could be found that the trends of refrigerant mass flow rates in the three circuits
during defrosting at Stages 1 and 2 are the same as those shown in Fig. 4.13. But at
Stage 3, the refrigerant mass flow rate of Circuit 1 decreased to 0 g/s at 172 s, as
designed in Table 4.2. At the same time, the values of the other two circuits increased
first and then kept fluctuating, with their same peak values at 12.59 g/s at 195 s into
defrosting. All their trends met the changes in the proportion of the refrigerant distri-
bution shown in Fig. 4.20B.