Determination of optimum refrigerant charge

for domestic refrigerator units

V. I. Dmitriyev and V. E. Pisarenko

Key words: refrigerant, refrigerant charge, domestic refrigerator

estimer /a quantit# optima/e. Des mesures exp~rimenta/es ant

D6termination de la charge #t# effectu~es avec diff#rentes charges qui ont montr# que /a
optimale en frigorig¢ne des temperature de/'air ambiant n "~tait pas un facteur important.
On a montr# que le coefficient de performance #tait plus
rdrig6rateurs m6nagers affect# par un exc#s que par un manque de charge. Un exc#s
de 10% de frigorigene r#duisait /e COP de 6 ~ 12% tandis
On indique une m~thode d~valuation de la charge qu'un manque de frigorig~ne du m&me ordre ne r#duisait le
optima/e en frigorigene (R 12) des r#frig#rateurs managers COP que de 4 ~ 6%.
uti/isant un d&tendeur a tube capi//aire. On a trouv# que /a On donne des #quations et des figures permettant
quantit# de frigorig#ne d#pendait surtout du volume de d ~va/uer la charge optima/e pour des r#frig#rateurs ayant des
/'~ vaporateur et du condenseur ~ peu pros en accord a vec une volumes int#rieurs de 120 ~ 180 dm 3 et des volumes de
analyse th6orique pr#c#dente qui tendait cependant ~ sur- condenseur et d'#vaporateur de/'ordre de 90 b 160 cm a.

A method is given of assessing the optimum tended to overestimate the optimum quantity.
refrigerant (R 12) charge for domestic refrige- Experimental measurements were made with
rators employing a capillary tube expansion different charges and it was concluded that
device. This amound of refrigerant was found to ambient air temperature was not a significant
depend mostly on the volumes of the evaporator factor. The coefficient of performance was
and condenser in broad accordance with an shown to be more sensitive to overfilling than
earlier theoretical analysis which, however, underfilling.

The peculiarities of operation of cooling units of In the first stage the optimum charge of refrigerant
domestic refrigerators with capillary tubes as the re- (R 12) for the cooling unit was found for a series of
gulating device cause difficulties in precisely calculat- evaporators with different capacities but practically the
ing the optimum charge of refrigerant. At present it is same heat transfer surface. A cooling unitwas mounted
determined experimentally during the testing of newly in a refrigerator with a known heat inflow and tested
developed models, but this takes too much labour and with different refrigerant charges, different fractional
time. operating times, and temperatures of the surrounding
The authors have developed a calculation method air of 25°C and 32°C.
for determining the optimum refrigerant charge. A The temperatures in the cabinet and of the sur-
preliminary theoretical analysis showed that this rounding air were measured by means of copper-
magnitude is mainly influenced by the inside volume of constantan thermocouples. The maximum error of
the system and particularly the capacities of the evap- measuring the temperature was -I-0.02°C. The power
orator and condenser. It is corroborated by the equa- consumed by the compressor was measured by a watt-
tion for the charge of refrigerant for commercial meter the error of measurements being +1.5W. The
refrigerating units, suggested by Anymark and coefficient of performance of the unit is defined by the
Rollsgord. ~ formula
G , = 2 0 + 0.6Ve (1) K c= O 0

where G, is the dose of refrigerant, g; and Ve is the inner where Q0 is the cold producton of unit, Oo=kF(ts-- tt) ; kF
capacity of evaporator, cm 3. is the heat inflow coefficient of cabinet, W~C - 1; ts is the
Experience of applying this formula to units of temperature of surrounding medium, °C; t r is the
domestic refrigerators showed that it gave over-filling average internal temperature of the cabinet °C, ie
This formula does not take sufficient account of the it= (tlV 1-~-t2V2)/V 1-F V2); qG are the average tempera-
peculiarities of the capillary tube and the condenser, so tures in low temperature and high temperature
a special investigation was necessary. Experiments sections, respectively, °C; V~, V2 are the capacity of low
have been conducted in two stages. temperature and high temperature sections respec-
tively, m3; N is the power consumed by compressor,
W: and b is the fractional operating time. When
The authors are at Politechnic Institute, Lenin Ave 168, Kishinev,
277004 USSR. Paper received 23 December 1982. defining Kc the maximum error was within +2%.

0140-7007/84/0031 78-0353.00
178 @ 1984 Butterworth S~ Co (Publishers) Ltd and IIR Revue Internationale du Froid
~° ~o. ~ ~0
o.4i,5o 65' 8o' ,'5 °4';5 ".g ;5 ,,"
• ' ~ Condenser capacity = 91cm5 Condenser capacity = il7cm 3
Ambient temperature = 250C Ambient temperature = 25°C

•" • O411 I I I/ O41^/ / ..1^ ./

• 65 80 95 I10 ' 80 95 I10 125
Gr,g Gr,~
Condenser capacity = 117cm3 Condenser capacity = 151crn3
O~1 I I I J/ o3L I I ( J| Ambient temperature = 32°C Ambient temperature =250C
65 80 95 I10 65 80 95 I10
er,g Grog Fig.3 Dependence of the coefficient of performance on refrigerant
Evaporator capacity = 139cm3 Evaporator capacity = 118cm3 charge with different condenser capacities for different fractional
Ambient temperature= 25°C Ambient temperature = 25% operating times ( ~ - 0 . 4 . 0-0.7 and ~ - 1 . 0 ) . Evaporator
volume= 1 39 cm 3
Fig.1 Effect of refrigerant charge on the coefficient of performance
for machines with different evaporation capacities, with different
fractional operating times ( n - 0 . 4 , 0 - 0 . 7 and A - 1 . 0 ) . Condenser Fig.3 Influence sur le COP de la charge de frigorig~ne avec
capacity= 117 cm 3 d/ff#rents vo/umes de condenseur pour d/ff#rents temps de fonct/on-
nement partiels ( > pourO,4, 4:pourO,7 et A pour 1,0). Volume de
Fig. 1 Influence de la charge de frigorig#ne sur le COP pour I~vaporateur: 139 cm 3
diff#rents volumes d'#vaporateur, diff#rents temps de fonctionne-
ment partials ([q pour 0.4, 0 pour 0,7 et A pour 1,0). Volume du Evaporator capacity, cm 5
condensuer, 117 cm 3 IOO IIO 120 1:30 140 150 160
I I

I00
f
90

90
80
O1

8O 70

60
70 I I I I I
1(30 I10 120 150 140
Vetcm3
50
Fig2 Dependence of the optimum charge on the evaporator
capacity (curve 1 Anvwarke and Rollsgord. Curve 2. o w n data)
40
Fig.2 Influence du volume de I'~vaporateur sur la charge optimale
(courbe 1, Anywarke et Rollsgord, courbe 2, r~sultats des auteurs)
50

\
The volumes of different condensers and evap-
orators were found both by expanding a known 20
volume of air into them and measuring the pressure,
and by weighing the water required to fill them. The 10 ,I
error in finding the capacity of the evaporators and
condensers was -I- 3%.
The refrigerator cabinet used had a heat inflow of
1.6 W°C -1, the capillary tube had an inner diameter of
0.8 mm and allowed an air flow of 1.05 x 10 -4 m 3 s-] at
-IO
a pressure difference of 0.8 MPa. The evaporator had
an outer heat transfer surface F~0.64 m 2 and capa-
cities of 103, 118 and 1 39 cm3; the condensers had an I I I I I I I
90 I00 I10 120 130 140 150
outer heat transfer surface F~ 1.32 m 2 and capacities of
91,117, and 1 51 c m 3, These evaporator and condenser Condenser capacity,cm3
designs are typical of domestic refrigerator units with Fig.4 Nomogram for determining the optimum refrigerant charge
capacities of 0.1 2 to 0.18 m 3. Fig.4 Abaque pour la d#termination de la charge optimale en
The results of the first stage of testing are given in frigorigbne

Volume 7 Num6ro 3 Mai 1984 179

 Fig.1 which shows that the optimum charge of the The dependence of optimum refrigerant charge on
refrigerant does not depend on the temperature of the the capacity of condenser, at constant evaporator
surrounding medium or the fractional operating time at capacity of 1 39 cm 3, may be approximated within 5%
different capacities of the evaporator. by a straight line given by the equation:
Insufficient refrigerant causes the evaporator to be
incompletely filled so the boiling pressure is lowered, G*=O.62Vc+ 18 (3)
the vapour superheat at entry to the compressor is The equations (2) and (3) are true only at specific
increased and the coefficient of performance of the unit values of evaporator condenser capacities. A nomog-
is lowered. On the other hand, overfilling the unit with ram (Fig.4) has been constructed to define an optimum
refrigerant reduces the active surface of condenser, dose of refrigerant for refrigerating units with any
thus increasing the temperature and pressure of conde- combinations of capacities of evaporator and conde-
nsation, lowering cold production and decreasing the nser. In the upper part of the nomogram there is shown
coefficient of performance. the dependence G~=f(Ve) (see Fig. 2). The upper and
The rate of decrease of Kc with overfilling is higher lower parts of the nomogram are linked by common
than thatwith underfilling the system. Thus, exceeding axes.
the optimum charge of refrigerant by 10%, as can be An example is given of optimum refrigerant charge
seen in Fig.l, causes a decrease ofK c of 6 - 1 2%, whilst at Ve= 118 c m 3 and Vc= 105 c m 3. In this case Gr~= 76 g.
decreasing the charge of refrigerant by the same The optimum refrigerant charge for the unit of a
amount decreases Kc by only 4 - 6 % The dependence domestic refrigerator can be also calculated by the
of the optimum refrigerant charge G j on the capacity of formula:
evaporator Ve may be correlated by:
Gr*= 0.41 Ve+ 0.62 Vc- 38 (4)
G~= 38+ 0.41Ve (2)
The nomogram and formula are applicable over a range
This formula is true for a capacity of condenser 117 of evaporator capacities of 100-140 cm 3 and conde-
cm 3, The dependence (2) is shown in Fig.2, where it is nser capacities of 90-1 50 cm 3, for a range of surround-
compared to dependence (1). ing air temperatures of 25-32°C.
In the second set of tests, the optimum charge of
refrigerant for different condenser capacities was in-
vestigated. The results are shown in Fig.3 which also Reference
show that the optimum charge of refrigerant with
Shavra, V. M, Selection of capillary tube and analysis of
different condensers does not depend on the tempera- hermetic refrigeratingmachineoperationKholodilnaya Tekhnika
ture of surrounding air or the fractional operating time. 11 (1966) 21-25

180 International Journal of Refrigeration