Purdue University

Purdue e-Pubs
International Compressor Engineering Conference School of Mechanical Engineering

2016

Performance Analysis of a Reciprocating

Compressor under Typical Transients of
Refrigeration Systems
Marco Carrilho Diniz
POLO - UFSC, Brazil, [email protected]

Cesar Jose Deschamps

POLO - UFSC, Brazil, [email protected]

Follow this and additional works at: https://docs.lib.purdue.edu/icec

Diniz, Marco Carrilho and Deschamps, Cesar Jose, "Performance Analysis of a Reciprocating Compressor under Typical Transients of
Refrigeration Systems" (2016). International Compressor Engineering Conference. Paper 2448.
https://docs.lib.purdue.edu/icec/2448

This document has been made available through Purdue e-Pubs, a service of the Purdue University Libraries. Please contact [email protected] for
additional information.
Complete proceedings may be acquired in print and on CD-ROM directly from the Ray W. Herrick Laboratories at https://engineering.purdue.edu/
Herrick/Events/orderlit.html
 1330 Page 1

Performance Analysis of a Reciprocating Compressor under Typical Transients of

Refrigeration Systems

Marco C. DINIZ1, Cesar J. DESCHAMPS1*

1
POLO Research Laboratories for Emerging Technologies in Cooling and Thermophysics
Federal University of Santa Catarina
Florianopolis, SC, Brazil
[email protected]

* Corresponding Author

ABSTRACT
The performance of reciprocating compressors is usually evaluated under steady-state operating conditions defined
in standards. However, the compressor is submitted to quite different conditions in actual refrigeration systems, such
as transients associated with the on/off cycling operation. This paper presents a simulation model developed to
evaluate the performance of reciprocating compressors during typical transients of refrigeration systems. The model
is formed by the coupling of two sub models, one for the compression chamber and the other to calculate the
temperatures and mass flow rates in components other than the compression chamber. As the model only simulates
the compressor, experimental data was required to provide input from the remaining components of the system. The
model was validated by comparing predictions and experimental data of discharged mass flow rate and power
consumption of a compressor operating under two transient system conditions. Then, the simulation model was
employed to predict the variation of the compressor efficiencies and temperatures during the typical transients
analyzed.

1. INTRODUCTION
About 1.4 billion household refrigerators and freezers are in use worldwide (Barther and Götz, 2012) and virtually
all of these refrigerators operate based on the vapor compression refrigeration cycle. The performance of these
appliances depends on the efficiency of each of its components, but also on the manner they interact with each other.

Despite the advent of variable speed compressors, the vast majority of the compressors employed on refrigerators
are still single speed. In this case, the refrigerator operation is characterized by alternate periods at which the
compressor is either operating (on) or idle (off). The analysis of the transient operating conditions of refrigerators
has been the subject of several researches during the last 30 years. However, most of such studies have focused their
attention either on the heat exchangers or on the throttling device.

Regarding the compressor operation during the refrigerator transients, two main aspects must the addressed. First,
and most important, is the variation of the pressure ratio to which the compressor is submitted. Unlike the conditions
during the refrigerator pulldown, when the compressor is started in a typical cycling condition of the refrigerator, the
time to reach a stabilized pressure ratio is usually short. However, this may differ from refrigerator to refrigerator
due to aspects such as the volume of the refrigerated compartment, volume of the heat exchangers, compressor
displacement volume, throttling device and refrigerant charge. The second aspect is related to the transient behavior
of the compressor temperatures during the on/off cycling conditions, which may vary even when the pressure ratio is
already stabilized.

Few studies have been developed regarding the compressor performance under transient refrigerator conditions.
Some of these studies focused on the crankshaft mechanism and/or bearing system during the first seconds after

23rd International Compressor Engineering Conference at Purdue, July 11-14, 2016

 1330 Page 2
startups and shutdowns (Dufour et al., 1995; Link and Deschamps, 2011). In other studies, the objective was to
predict mass flow rate and power consumption (Porkhial et al., 2002; Hermes and Melo, 2008; Ndiaye and Bernier,
2010; Negrao et al., 2011) or heat transfer phenomena (Lohn et al., 2015). None of such works has focused on
understanding the phenomena that influence the efficiency of the compressor during typical transients of
refrigeration systems.

This paper presents a simulation model developed to characterize the performance of reciprocating compressors
under transient system conditions. It consists of two sub models: one to predict the compression chamber and the
other to calculate the variation of the temperatures and mass flow rates in components other than the compression
chamber. Some input data required for the model were obtained from measurements, which were also used to
validate the model predictions.

2. EXPERIMENTAL PROCEDURE
The model developed in this work only simulates the compressor. Therefore, the input parameters that would be
calculated by models for the other components of the system had to be experimentally obtained. The required
parameters are the evaporating and condensing pressures and the gas temperature at the compressor inlet.
Additionally, several parameters were also measured in the experiments to validate model predictions, including the
mass flow rate, indicated power, electrical power consumption and compressor temperatures.

The tested system was a one door typical frost-free 300-L refrigerator, in which the compressor is on-off controlled
by the freezer temperature, while a mechanical damper is responsible to establish the temperature in the fresh food
compartment. The original compressor employed in the refrigerator was substituted by an instrumented compressor
with 30% less volumetric displacement than the original and the refrigerant charge (R600a) was set to 27g. The tests
were performed with the refrigerator in a controlled environment with temperature of 25ºC and 50% of relative
humidity. The refrigerator temperature setting was set to its medium value, at which the freezer temperature reached
-16ºC by the time the compressor was switched off.

Two different “on” periods were analyzed in this study. The first was typical of the periodic cyclic operation of the
refrigerator. In this case, the equalizing pressure was around 0.8 bar and the “on” cycle had a duration of around 15
minutes. The second period, usually termed recovery on period, takes place immediately after refrigerator defrost. In
this case, the equalizing pressure before the compressor is switched on was around 2.6 bar and the “on” period
presented a duration of approximately 60 minutes. The transient pressures and suction line temperature that were
used as input for the simulation model will be presented and discussed further.

3. MATHEMATICAL MODEL

The model consists of two sub models: one to simulate the compression chamber and the other to calculate the
variation of compressor temperatures and mass flow rates. The time scale of the compression cycle (milliseconds) is
considerably smaller than that of the compressor thermal transients (minutes), allowing the compression chamber to
be simulated using a quasi-steady state approach. The coupled solution procedure will be detailed in section 3.3.

3.1 Compression chamber

The model for the compression chamber follows a transient lumped formulation based on four groups of equations
(Todescat et al., 1992). The first group allows the calculation of the instantaneous volume occupied by the gas inside
the compression chamber. The second group is the conservation equations applied to the compression chamber. The
mass conservation equation is given by:

𝑑𝑚𝑔
= 𝑚̇𝑠𝑣 − 𝑚̇𝑠𝑣,𝑏 − 𝑚̇𝑑𝑣 + 𝑚̇𝑑𝑣,𝑏 − 𝑚̇𝑝𝑐 (1)
𝑑𝑡

In Equation (1), 𝑚̇𝑠𝑣 and 𝑚̇𝑑𝑣 are the instantaneous mass flow rates through the suction and discharge valves, while
𝑚̇𝑝𝑐 represents the instantaneous leakage through the piston cylinder gap and the subscript 𝑏 indicates backflow in

23rd International Compressor Engineering Conference at Purdue, July 11-14, 2016

 1330 Page 3
both valves. With the instantaneous mass and volume of refrigerant inside the compression chamber it is possible to
calculate its density.

The instantaneous temperature of the refrigerant inside the compression chamber is obtained by applying the energy
conservation equation, given by:

𝑑𝑇𝑔
= 𝐴 − 𝐵𝑇𝑔 (2)
𝑑𝑡

where

1 𝑑𝑚𝑔
𝐴= [𝐻𝑔 𝐴𝑤 𝑇𝑤 − ℎ𝑔 + (𝑚̇𝑠𝑣 ℎ𝑠𝑐 ) − (𝑚̇𝑠𝑣,𝑏 ℎ𝑔 ) − (𝑚̇𝑑𝑣 ℎ𝑔 ) + (𝑚̇𝑑𝑣,𝑏 ℎ𝑑𝑐 ) − (𝑚̇𝑝𝑐 ℎ𝑔 )] (3)
𝑚𝑔 𝑐𝓋 𝑑𝑡

1 𝜕𝑝𝑔 𝑑∀𝑐𝑐 𝜕𝑝𝑔 𝑑𝑚𝑔

𝐵= [𝐻𝑔 𝐴𝑤 + | − | 𝓋𝑔 ] (4)
𝑚𝑔 𝑐𝓋 𝜕𝑇𝑔 𝑑𝑡 𝜕𝑇𝑔 𝑑𝑡
𝓋 𝓋

In the above equations 𝐻𝑔 is a convective heat transfer coefficient calculated using the correlation proposed by
Annand (1963). The method to estimate the mass flow rates in the above equations will be explained shortly. The
temperatures of the suction chamber (𝑇𝑠𝑐 ), discharge chamber (𝑇𝑑𝑐 ) and cylinder wall (𝑇𝑤 ) are calculated by the
model that will be described in section 3.2. With the temperature calculated using Equation (2) and the density of the
refrigerant it is possible to estimate the pressure inside the compression chamber by means of a state equation.

The third group of equations is used to estimate the dynamics of the suction and discharge valves. A single degree-
of-freedom mass-spring model is employed, with effective force areas being used to characterize the force acting on
the valve. Finally, the fourth group of equations predicts the mass flow rates that cross the control volume surface.
Mass flow rate through the valves is estimated using the concept of effective flow area to correct the theoretical
model for the isentropic flow through a convergent nozzle. Leakage through the piston cylinder gap is estimated by
considering a fully developed laminar Couette-Poiseuille flow (Ferreira and Lilie, 1984).

The parameters associated with the compression chamber that are used as input in the other sub model are calculated
as shown in Table 1. The mechanical losses 𝑄̇𝑏 and the electrical efficiency 𝜂𝑒𝑙𝑒 were specified based on
experimental data previously obtained for the compressor under analysis.

Table 1 Output from the compression chamber model.

Parameter Calculation

𝑊̇𝑖𝑛𝑑 𝑓 ∮ 𝑝𝑔 𝑑∀𝑐𝑐

𝑚̇𝑠 , 𝑚̇𝑑 , 𝑚̇𝑙 , 𝑚̇𝑠,𝑏

𝑓 ∮ 𝑚̇𝑖 𝑑𝑡, where 𝑖 = 𝑠𝑣, 𝑖 = 𝑑𝑣, 𝑖 = 𝑝𝑐, 𝑖 = 𝑠𝑣, 𝑏 or 𝑖 = 𝑑𝑣, 𝑏
and 𝑚̇𝑑,𝑏
1
𝑇𝑑 ∮ 𝑚̇𝑑𝑣 𝑇𝑔 𝑑𝑡
𝑚̇𝑑
1
𝑇𝑙 ∮ 𝑚̇𝑝𝑐 𝑇𝑔 𝑑𝑡
𝑚̇𝑙
1
𝑇𝑠,𝑏 ∮ 𝑚̇𝑠𝑣,𝑏 𝑇𝑔 𝑑𝑡
𝑚̇𝑠,𝑏
𝑄̇𝑚𝑜𝑡 (𝑊̇𝑖𝑛𝑑 + 𝑄̇𝑏 )/𝜂𝑒𝑙𝑒 − (𝑊̇𝑖𝑛𝑑 + 𝑄̇𝑏 )

23rd International Compressor Engineering Conference at Purdue, July 11-14, 2016

 1330 Page 4

3.2 Temperatures and mass flow rates through control volumes

To calculate the temperature and mass flow rates inside components other than the compression chamber, transient
energy and mass balances were applied to the control volumes presented in Figure 1. The energy balances applied to
each control volume are presented in Table 2 in the Appendix. The mass balances can also be deduced through the
equations presented in Table 2.

Figure 1: Control volumes for the thermal analysis.

In the compressor employed in this work, part of the gas in the suction line flows directly to the suction muffler,
while the other share fills the internal environment (Figure 1). In the model, this flow bifurcation is represented by a
factor 𝜑. The 𝜑 factor and the heat conductances between the components were obtained by adjusting the model to a
measured steady state temperature distribution in a calorimeter, a procedure that has been previously adopted by
several other authors, such as Todescat et al. (1992). Naturally, the transient terms in the system of equations were
not taken into account in this adjustment procedure. The condition chosen for the adjustment consisted on an
evaporating temperature 𝑇𝑒 = −28 °𝐶 and a condensing temperature 𝑇𝑐 = 40 °𝐶 with ambient temperature 𝑇𝑎𝑚𝑏 =
25 °C. The adjusted heat conductances and the 𝜑 factor were maintained constant throughout the simulation of the
compressor under transient conditions.

The nonlinear system of equations represented in Table 2 is solved by means of a Newton-Raphson method. Of
especial interest are the suction chamber, cylinder and discharge chamber temperatures, which are used as input data
in the compression chamber model.

3.3 Solution procedure

The solution procedure is based on exchange of information between the two previously discussed models. Figure 2
presents the flowchart that characterizes the coupled solution for the overall model. The instantaneous operating
conditions are given by the system condensing and evaporating pressures and compressor inlet temperature,
obtained from the experimental tests of the refrigerator.

23rd International Compressor Engineering Conference at Purdue, July 11-14, 2016

 1330 Page 5
Since the compression chamber and the thermal model require input data from each other, inside each time step an
iterative procedure is necessary to verify convergence, which is considered to be reached when the suction chamber
temperature and the inlet mass flow rate present a variation between successive iterations of less than 0.1%.

Start (t=0)

Specify initial values for compressor

temperatures and guesses mass flows

Define instantaneous
operating conditions

Solve compression cycle model

Solve energy balances to update

compressor temperatures t = t + Δt

Solve mass balances to update

mass flow rates

No
Convergence criteria
satisfied?

Yes

No
t > t F?

Yes

End of simulation (t=t F)

Figure 2: Flowchart of the solution procedure.

4. RESULTS
Figures 3 and 4 present the measured suction and discharge pressures during the two “on” periods considered in this
study. The data is normalized by the total time of each period (15 min. and 60 min.). It can be observed that for the
typical cycling condition the suction pressure rapidly approaches its stabilized value. On the other hand, the transient
associated with the defrost recovery is quite different, and resembles more a refrigerator pulldown.

The transient evaporating and condensing pressures shown in Figures 3 and 4 and the compressor inlet temperature
were used as input data for the compressor model. Figures 5 and 6 show the comparison between experimental and
numerical results of mass flow rate in the two transients (cycling and defrost recovery), with good agreement being
observed for both cases. The discrepancy observed during the first 20 to 30 seconds, only visible in Figure 5, cannot
be fully explained at this stage of our study but is more likely to be a result of measurement errors due to the
presence of some liquid droplets inside the measurement section of the flow meter. Considering the cycling
condition, the results show that the variation of mass flow rate is not significant after a normalized time around 0.2.
On the other hand, this variation of mass flow rate is slightly more intense for the defrost recovery condition due to
the variation of the suction pressure, as shown in Figure 4.

Figures 7 and 8 present the evolution of electrical power consumption and indicated power during the typical
cycling and defrost recovery conditions. As observed for the mass flow rate, the trends of variation are different,
with the defrost recovery showing a considerable variation during the whole period. Some discrepancies between

23rd International Compressor Engineering Conference at Purdue, July 11-14, 2016

 1330 Page 6
predictions and measurements can be observed for the indicated power, which can partially be attributed to absence
of models to predict the pressure pulsations in the suction and discharge systems, as well as to errors in the leakage
and heat transfer models. Furthermore, the more significant discrepancies observed for the electrical power
consumption can be attributed to lack of appropriate models for the mechanical losses and electrical motor
efficiency, but also to the aforementioned discrepancies regarding the indicated power.

1 6 1 7

6
5

Evaporating pressure [bar]

Evaporating pressure [bar]

Condensing pressure [bar]

Condensing pressure [bar]
0.9 0.9
5
4
0.8 Evaporating 0.8 Evaporating 4
3
Condensing Condensing 3
0.7 0.7
2
2
0.6 0.6
1 1

0.5 0 0.5 0
0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1
Time [-] Time [-]
Figure 3: Transient variation of evaporating and Figure 4: Transient variation of evaporating and
condensing pressures for the typical cycling condensing pressures for the defrost recovery

2 2
Mass Flow Rate [kg/h]

Experimental
Mass Flow Rate [kg/h]

Experimental

Numerical Numerical
1.5 1.5

1 1

0.5 0.5
0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1
Time [-] Time [-]
Figure 5: Numerical and experimental results of mass Figure 6: Numerical and experimental results of mass
flow rate for the typical cycling flow rate for the defrost recovery

55 55

Experimental 50 Electrical power Experimental

50
Electrical power Numerical Numerical
45 45
Power [W]
Power [W]

40 40
Indicated power
35 Indicated power 35

30 30

25 25
0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1
Time [-] Time [-]
Figure 7: Numerical and experimental results of Figure 8: Numerical and experimental results of
electrical and indicated power for the typical cycling electrical and indicated power for the defrost recovery

23rd International Compressor Engineering Conference at Purdue, July 11-14, 2016

 1330 Page 7
Figure 9 and 10 show the transient variation of the temperatures in suction chamber and at the cylinder wall. Despite
the deviations, the trends are well captured by the simulation model, which predicts the temperatures in both
conditions with an average error of around 2ºC. In part, the deviations can be attributed to the fact that the heat
conductances and the 𝜑 factor were maintained constant throughout the transient simulation. These coefficients are
expected to vary, especially during the first minutes of compressor operation, due to the considerable variation in the
mass flow rate. Therefore, improvements in this methodology will be incorporated in future work.

65 65
Cylinder Cylinder
60 60

55 55

Temperature [ºC]
Temperature [ºC]

Experimental Experimental
50 50
Numerical Numerical
45 45

40 40

35 35 Suction chamber
Suction chamber
30 30
0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1
Time [-] Time [-]

Figure 9: Numerical and experimental results of Figure 10: Numerical and experimental results of
temperatures in the suction chamber and at the cylinder temperatures in the suction chamber and at the cylinder
wall for the typical cycling wall for the defrost recovery

Figures 11 and 12 presents the pressure ratio to which the compressor is submitted and the numerical results of
thermodynamic, volumetric and global the efficiencies for the two conditions analyzed. For the cycling condition
(Figure 11), the thermodynamic and global efficiencies approach a minimum value in the beginning of the period
due to aspects such as suction superheating and losses in the suction and discharge processes. A local minima for
these efficiencies is also observed in the defrost recovery moments after the compressor is turned on. However, the
global minima for this condition are observed immediately before compressor shut down, mainly as a result of the
increase in compressor temperatures. On the other hand, for both conditions, the volumetric efficiency presents a
monotonic behavior that is mainly governed by the pressure ratio.

1 1 1 1
Thermodynamic Thermodynamic
0.9 0.8 0.9 0.8

0.6
Efficiencies [-]

0.6
Efficiencies [-]

0.8 0.8 Volumetric

pe / pc [-]

pe / pc [-]
Volumetric

0.7 0.4 0.7 0.4

0.6 Global 0.2 0.6 0.2

Global

Pressure ratio Pressure ratio

0.5 0 0.5 0
0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1
Time [-] Time [-]
Figure 11: Numerical results of efficiencies for the Figure 12: Numerical results of efficiencies for the
typical cycling defrost recovery

Figures 13 and 14 present the effect of suction superheating (∆𝑇𝑠𝑢𝑝 ) on the thermodynamic power (𝜙) and mass
flow rates (𝜓) during both transients. The peak of suction superheating reached soon after the compressor is
switched on is more intense for the cycling condition. This increases the power necessary the compress the gas by
almost 10%, which explains in part the minimum value for the thermodynamic efficiency (Figure 11). The mass

23rd International Compressor Engineering Conference at Purdue, July 11-14, 2016

 1330 Page 8
flow rate is also reduced from its theoretical value, but this impact is smaller than that due to the pressure ratio. The
results for defrost recovery condition show that the negative effect of the suction superheating increases with time.

12 30 12 30
10 10

Impact on theoretical value [%]

Impact on theoretical value [%]

8 25 8 25
6 6

Superheating [ºC]
Superheating [ºC]
20 20
4 4
2 2
15 15
0 0
-2 -2
10 10
-4 -4
-6 5 -6 5
-8 -8
-10 0 -10 0
0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1
Time [-] Time [-]
Figure 13: Impact of suction superheating during the Figure 14: Impact of suction superheating during the
cycling on recovery on

6. CONCLUSIONS
This paper presented a simulation model to calculate the performance of reciprocating compressors under two
transient conditions found in refrigerators, represented by standard cycling and defrost recovery conditions. The
model consists of two sub models, one for the compression chamber and the other to predict the compressor
temperature distribution and mass flow rates through components other than the compression chamber. A
compressor was tested in a refrigerator and several parameters were measured (mass flow rate, indicated power,
electrical energy consumption, temperatures). Some of these measurements were used as input data and others to
validate the simulation model. Overall, the model was shown to be capable of predicting the transient performance
of a compressor in a refrigerator. The results showed that the volumetric efficiency, which is closely related to the
pressure ratio, presents a monotonic variation during both transient transients. The thermodynamic and global
efficiencies presented a minimum value in the typical cycling condition immediately after the compressor start up.
This minimum value was shown to be strongly related to the suction gas superheating. For the recovery on the
minimum value of thermodynamic and global efficiencies were found to occur immediately before compressor
shutdown.

NOMENCLATURE
General symbols
𝐴 Area [m2] 𝑄̇𝑚𝑜𝑡 Motor losses [W]
𝑓 Compressor speed [Hz] 𝑇 Temperature [ºC]
𝐻 Heat transfer coefficient [W/m2K] 𝑇𝑑 Average discharge temperature [ºC]
ℎ Enthalpy [J/kgK] 𝑈𝐴 Heat conductance [W/K]
𝑚̇ Mass [kg] ∀ Volume [m3]
𝑚̇ Mass flow rate [kg/s] 𝑊̇𝑖𝑛𝑑 Indicated power [W]
𝑚̇𝑑 Mass flow rate at discharge orifice [kg/s] 𝜂𝑔 Global efficiency [-]
𝑚̇𝑠 Mass flow rate at suction orifice [kg/s] 𝜂𝑡 Thermodynamic efficiency [-]
𝑝 Pressure [Pa] 𝜂𝑣 Volumetric efficiency [-]
𝑄̇ Heat transfer rate [W] 𝜑 Semi-direct suction factor [-]
𝑄̇𝑏 Mechanical losses [W]
Subscripts
𝑎𝑚𝑏 External environment 𝑙 Leakage
𝑏 Related to backflow 𝑚𝑜𝑡 Electrical motor
𝑐 Condensing 𝑜𝑖𝑙 Lubricant oil

23rd International Compressor Engineering Conference at Purdue, July 11-14, 2016

 1330 Page 9
𝑐𝑐 Compression chamber 𝑝𝑐 Piston cylinder gap
𝑑 Averaged value at discharge orifice 𝑠 Averaged value at suction orifice
𝑑𝑐 Discharge chamber 𝑠𝑐 Suction chamber
𝑑𝑙 Discharge line 𝑠𝑒 Suction muffler inlet
𝑑𝑚 Discharge muffler 𝑠ℎ𝑒𝑙𝑙 Compressor shell
𝑑𝑜 Discharge muffler outlet 𝑠𝑙 Suction line
𝑑𝑡 Discharge tube 𝑠𝑚 Suction muffler
𝑑𝑣 Discharge valve 𝑠𝑣 Suction valve
𝑒 Evaporating 𝑡ℎ Theoretical value
𝑖𝑒 Compressor internal environment 𝑤 Compression chamber wall
𝑔 Gas inside compression chamber

REFERENCES
Annand, W. J. D. (1963). Heat transfer in the cylinders of reciprocating internal combustion engines. Proc. I. Mech.
Eng. 117 973-96.

Barther, C. and Götz, T. (2012). The overall worldwide saving potential from domestic refrigerators and freezers.
Wuppertal, Germany. Wuppertal Institute for Climate, Environment and Energy.

Dufour, R., Hagopian, J. D., & Lalanne, M. (1995). Transient and steady state dynamic behavior of single cylinder
compressors: prediction and experiments. J. Sound Vib., 181(1), 23-41.

Hermes, C. J. L. & Melo, C. (2008). A first-principles simulation model for the start-up and cycling transients of
household refrigerators. Int. J. Refrig., 31, 1341-1357.

Ferreira, R. T. S. & Lilie, D. E. B. (1984). Evaluation of the leakage through the clearance between piston and
cylinder in hermetic compressor. In: Proc. Int. Compressor Engineering Conference at Purdue University, (paper
424). West Lafayette, USA.

Lohn, S. K., Diniz, M. C. & Deschamps, C. J. (2015). A thermal model for analysis of hermetic reciprocating
compressors under the on-off cycling operating condition. IOP. Conf. Ser.: Mater. Sci. Eng., 90, 012068.

Link, R., & Deschamps, C. J. (2011). Numerical modeling of startup and shutdown transients in reciprocating
compressors. Int. J. Refrig., 34, 1398-1414.

Ndiaye, D. & Bernier, M. (2010). Dynamic model of a hermetic reciprocating compressor in on–off cycling
operation (Abbreviation: Compressor dynamic model). Appl. Therm. Eng., 30, 792-799.

Negrão, C. O. R., Erthal, R. H., Andrade, D. E. V. & Silva, L. W. (2011). A semi-empirical model for the unsteady-
state simulation of reciprocating compressors for household refrigeration applications. Appl. Therm. Eng., 31, 1114-
1124.

Phorkial, S., Khastoo & Razavi, M. R. M. (2002). Transient characteristic of reciprocating compressors in household
refrigerators. Appl. Therm. Eng., 22, 1391-1402.

Todescat, M. L., Fagotti, F., Prata, A. T. & Ferreira, R. T. S. (1992). Thermal energy analysis in reciprocating
hermetic compressors. In: Proc. Int. Compressor Engineering Conference at Purdue University, (paper 936). West
Lafayette, USA.

ACKNOWLEDGEMENT
The present study was developed as part of a technical-scientific cooperation program between the Federal
University of Santa Catarina and EMBRACO. The authors also acknowledge the support of EMBRAPII Unit
POLO/UFSC and CAPES (Coordination for the Improvement of High Level Personnel).

23rd International Compressor Engineering Conference at Purdue, July 11-14, 2016

 1330 Page 1

APPENDIX

Table 2 Energy balances of thermal hydraulic sub model

𝒅
𝑪𝑽 (𝒎𝒖) ∑ 𝒎̇𝒉|𝒊𝒏 ∑ 𝒎̇𝒉|𝒐𝒖𝒕 ∑ 𝑸̇ ∑ 𝑾̇
𝒅𝒕
1 − 𝑚̇𝑖𝑒 ℎ𝑖𝑒 + (1 − 𝜑)𝑚̇𝑠𝑙 ℎ𝑠𝑙 𝑚̇𝑠𝑒 ℎ𝑠𝑒 − −
𝑡+Δ𝑡 𝑡 𝑡+Δ𝑡 𝑡
𝑢𝑠𝑚 − 𝑢𝑠𝑚 𝑚𝑠𝑚 − 𝑚𝑠𝑚
2 𝑚𝑠𝑚 + 𝑢𝑠𝑚 𝑚̇𝑠𝑒 ℎ𝑠𝑒 + 𝑚̇𝑠,𝑏 ℎ𝑠,𝑏 + 𝑚̇𝑙 ℎ𝑖𝑒 𝑚̇𝑠 ℎ𝑠𝑐 −𝑈𝐴𝑖𝑒−𝑠𝑚 (𝑇𝑖𝑒 − 𝑇𝑠𝑚 ) −
Δ𝑡 Δ𝑡
𝑇𝑤𝑡+Δ𝑡 − 𝑇𝑤𝑡
3 𝑚𝑤 𝑐𝑝𝑤 𝑚̇𝑠 ℎ𝑠𝑐 + 𝑚̇𝑑,𝑏 ℎ𝑑𝑐 𝑚̇𝑠,𝑏 ℎ𝑠,𝑏 + 𝑚̇𝑑 ℎ𝑑 + 𝑚̇𝑙 ℎ𝑙 𝑈𝐴𝑤−𝑜𝑖𝑙 (𝑇𝑤 − 𝑇𝑜𝑖𝑙 ) 𝑊̇𝑖𝑛𝑑 + 𝑄̇𝑏
Δ𝑡
𝑡+Δ𝑡 𝑡 𝑡+Δ𝑡 𝑡
𝑢𝑑𝑐 − 𝑢𝑑𝑐 𝑚𝑑𝑐 − 𝑚𝑑𝑐
4 𝑚𝑑𝑐 + 𝑢𝑑𝑐 𝑚̇𝑑 ℎ̅𝑑 𝑚̇𝑑𝑐 ℎ𝑑𝑐 + 𝑚̇𝑑,𝑏 ℎ𝑑𝑐 𝑈𝐴𝑑𝑐−𝑖𝑒 (𝑇𝑑𝑐 − 𝑇𝑖𝑒 ) −
Δ𝑡 Δ𝑡
𝑡+Δ𝑡 𝑡 𝑡+Δ𝑡 𝑡
𝑢𝑑𝑚 − 𝑢𝑑𝑚 𝑚𝑑𝑚 − 𝑚𝑑𝑚
5 𝑚𝑑𝑚 + 𝑢𝑑𝑚 𝑚̇𝑑𝑐 ℎ𝑑𝑐 𝑚̇𝑑𝑚 ℎ𝑑𝑜 𝑈𝐴𝑑𝑚−𝑖𝑒 (𝑇𝑑𝑚 − 𝑇𝑖𝑒 ) −
Δ𝑡 Δ𝑡
𝑡+Δ𝑡 𝑡 𝑡+Δ𝑡 𝑡
𝑢𝑑𝑡 − 𝑢𝑑𝑚 𝑚𝑑𝑡 − 𝑚𝑑𝑡
6 𝑚𝑑𝑡 + 𝑢𝑑𝑡 𝑚̇𝑑𝑚 ℎ𝑑𝑜 𝑚̇𝑑𝑙 ℎ𝑑𝑙 𝑈𝐴𝑑𝑡−𝑖𝑒 (𝑇𝑑𝑡 − 𝑇𝑖𝑒 ) −
Δ𝑡 Δ𝑡
𝑡+Δ𝑡
𝑇𝑚𝑜𝑡 − 𝑇𝑚𝑜𝑡 𝑡 𝑈𝐴𝑚𝑜𝑡−𝑖𝑒 (𝑇𝑚𝑜𝑡 − 𝑇𝑖𝑒 )
7 𝑚𝑚𝑜𝑡 𝑐𝑝𝑚𝑜𝑡 − − − 𝑄𝑚𝑜𝑡 −
Δ𝑡
𝑡+Δ𝑡 𝑡
𝑇𝑜𝑖𝑙 − 𝑇𝑜𝑖𝑙 𝑈𝐴𝑜𝑖𝑙−𝑠ℎ𝑒𝑙𝑙 (𝑇𝑜𝑖𝑙 − 𝑇𝑠ℎ𝑒𝑙𝑙 )
8 𝑚𝑜𝑖𝑙 𝑐𝑝𝑜𝑖𝑙 − − −
Δ𝑡 − 𝑈𝐴𝑤−𝑜𝑖𝑙 (𝑇𝑤 − 𝑇𝑜𝑖𝑙 )
𝑡+Δ𝑡 𝑡 𝑈𝐴𝑠ℎ𝑒𝑙𝑙−𝑎𝑚𝑏 (𝑇𝑠ℎ𝑒𝑙𝑙 − 𝑇𝑎𝑚𝑏 )
𝑇𝑠ℎ𝑒𝑙𝑙 − 𝑇𝑠ℎ𝑒𝑙𝑙
9 𝑚𝑠ℎ𝑒𝑙𝑙 𝑐𝑝𝑠ℎ𝑒𝑙𝑙 − − − 𝑈𝐴𝑜𝑖𝑙−𝑠ℎ𝑒𝑙𝑙 (𝑇𝑜𝑖𝑙 − 𝑇𝑠ℎ𝑒𝑙𝑙 ) −
Δ𝑡 − 𝑈𝐴𝑖𝑒−𝑠ℎ𝑒𝑙𝑙 (𝑇𝑖𝑒 − 𝑇𝑠ℎ𝑒𝑙𝑙 )

𝑈𝐴𝑖𝑒−𝑠ℎ𝑒𝑙𝑙 (𝑇𝑖𝑒 − 𝑇𝑠ℎ𝑒𝑙𝑙 )

+ 𝑈𝐴𝑖𝑒−𝑠𝑚 (𝑇𝑖𝑒 − 𝑇𝑠𝑚 )
𝑡+Δ𝑡 𝑡 𝑡+Δ𝑡 𝑡
𝑢𝑖𝑒 − 𝑢𝑖𝑒 𝑚𝑖𝑒 − 𝑚𝑖𝑒 − 𝑈𝐴𝑑𝑐−𝑖𝑒 (𝑇𝑑𝑐 − 𝑇𝑖𝑒 )
10 𝑚𝑖𝑒 + 𝑢𝑖𝑒 𝜑𝑚̇𝑠𝑙 ℎ𝑠𝑙 + 𝑚̇𝑙 ℎ𝑙 𝑚̇𝑙 ℎ𝑙 + 𝑚̇𝑖𝑒 ℎ𝑖𝑒 −
Δ𝑡 Δ𝑡 − 𝑈𝐴𝑑𝑚−𝑖𝑒 (𝑇𝑑𝑚 − 𝑇𝑖𝑒 )
− 𝑈𝐴𝑑𝑡−𝑖𝑒 (𝑇𝑑𝑡 − 𝑇𝑖𝑒 )
− 𝑈𝐴𝑚𝑜𝑡−𝑖𝑒 (𝑇𝑚𝑜𝑡 − 𝑇𝑖𝑒 )

23rd International Compressor Engineering Conference at Purdue, July 11-14, 2016