Related Commercial Resources

CHAPTER 43

COMPONENT BALANCING IN REFRIGERATION SYSTEMS

Refrigeration System ..................................................................................................................... 43.1
Components................................................................................................................................... 43.1
Selecting Design Balance Points .................................................................................................. 43.2
Energy and Mass Balances ........................................................................................................... 43.3
System Performance...................................................................................................................... 43.4

T HIS chapter describes methods and components used in bal-

ancing a primary refrigeration system. A refrigerant is a fluid
used for heat transfer in a refrigeration system. The fluid absorbs
Generally, the smallest differential between saturated evaporator
and saturated condensing temperatures results in the lowest energy
requirement for compression. Liquid refrigerant cooling or subcool-
heat at a low temperature and pressure and transfers heat at a higher ing should be used where possible to improve efficiencies and min-
temperature and pressure. Heat transfer can involve either a com- imize energy consumption.
plete or partial change of state in the case of a primary refrigerant. Where intermediate pressures have not been specifically set for
Energy transfer is a function of the heat transfer coefficients; tem- system operation, the compressors automatically balance at their
perature differences; and amount, type, and configuration of the respective suction and discharge pressures as a function of their rel-
heat transfer surface and, hence, the heat flux on either side of the ative displacements and compression efficiencies, depending on
heat transfer device. load and temperature requirements. This chapter covers the tech-
nique used to determine the balance points for a typical brine chiller,
but the theory can be expanded to apply to single- and two-stage
REFRIGERATION SYSTEM
systems with different types of evaporators, compressors, and con-
A typical basic direct-expansion refrigeration system includes an densers.
evaporator, which vaporizes incoming refrigerant as it absorbs heat,
increasing the refrigerant’s heat content or enthalpy. A compressor COMPONENTS
pulls vapor from the evaporator through suction piping and com-
Evaporators may have flooded, direct-expansion, or liquid
presses the refrigerant gas to a higher pressure and temperature. The
overfeed cooling coils with or without fins. Evaporators are used to
refrigerant gas then flows through the discharge piping to a con-
cool air, gases, liquids, and solids; condense volatile substances; and
denser, where it is condensed by rejecting its heat to a coolant (e.g.,
freeze products.
other refrigerants, air, water, or air/water spray). The condensed liq-
Ice-builder evaporators accumulate ice to store cooling energy
uid is supplied to a device that reduces pressure, cools the liquid by
for later use. Embossed-plate evaporators are available (1) to cool a
flashing vapor, and meters the flow. The cooled liquid is returned to
falling film of liquid; (2) to cool, condense, and/or freeze out vola-
the evaporator. For more information on the basic refrigeration cycle,
tile substances from a fluid stream; or (3) to cool or freeze a product
see Chapter 1 of the 2005 ASHRAE Handbook—Fundamentals.
by direct contact. Brazed- and welded-plate fluid chillers can be
Gas compression theoretically follows a line of constant entropy.
used to improve efficiencies and reduce refrigerant charge.
In practice, adiabatic compression cannot occur because of friction
Ice, wax, or food products are frozen and scraped from some
and other inefficiencies of the compressor. Therefore, the actual
freezer surfaces. Electronic circuit boards, mechanical products, or
compression line deviates slightly from the theoretical. Power to the
food products (where permitted) are flash-cooled by direct immer-
compressor shaft is added to the refrigerant, and compression
sion in boiling refrigerants. These are some of the diverse applica-
increases the refrigerant’s pressure, temperature, and enthalpy.
tions demanding innovative configurations and materials that
In applications with a large compression ratio (e.g., low-
perform the function of an evaporator.
temperature freezing, multitemperature applications), multiple
Compressors can be positive-displacement, reciprocating-
compressors in series are used to completely compress the refriger-
piston, rotary-vane, scroll, single and double dry and lubricant-
ant gas. In multistage systems, interstage desuperheating of the
flooded screw devices, and single-stage or multistage centrifugals.
lower-stage compressor’s discharge gas protects the high-stage
They can be operated in series or in parallel with each other, in
compressor. Liquid refrigerant can also be subcooled at this inter-
which case special controls may be required.
stage condition and delivered to the evaporator for improved effi-
Drivers for compressors can be direct hermetic, semihermetic, or
ciencies.
open with mechanical seals on the compressor. In hermetic and
An intermediate-temperature condenser can serve as a cascading
semihermetic drives, motor inefficiencies are added to the refriger-
device. A low-temperature, high-pressure refrigerant condenses on
ant as heat. Open compressors are driven with electric motors, fuel-
one side of the cascade condenser surface by giving up heat to a low-
powered reciprocating engines, or steam or gas turbines. Intermedi-
pressure refrigerant that is boiling on the other side of the surface.
ate gears, belts, and clutch drives may be included in the drive.
The vapor produced transfers energy to the next compressor (or
Cascade condensers are used with high-pressure, low-
compressors); heat of compression is added and, at a higher pres-
temperature refrigerants (such as R-23) on the bottom cycle, and
sure, the last refrigerant is condensed on the final condenser surface.
high-temperature refrigerants (such as R-22, azeotropes, and refrig-
Heat is rejected to air, water, or water spray. Saturation temper-
erant blends or zeotropes) on the upper cycle. Cascade condensers
atures of evaporation and condensation throughout the system fix
are manufactured in many forms, including shell-and-tube, em-
the terminal pressures against which the single or multiple compres-
bossed plate, submerged, direct-expansion double coils, and brazed
sors must operate.
or welded plate heat exchangers. The high-pressure refrigerant
from the compressor(s) on the lower cycle condenses at a given
The preparation of this chapter is assigned to TC 10.1, Custom Engineered intermediate temperature. A separate, lower-pressure refrigerant
Refrigeration Systems. evaporates on the other side of the surface at a somewhat lower

43.1
Copyright © 2006, ASHRAE
43.2 2006 ASHRAE Handbook—Refrigeration (SI)

temperature. Vapor formed from the second refrigerant is com- entrance. If the thermodynamic pressure requirement becomes too
pressed by the higher-cycle compressor(s) until it can be condensed great for a given speed and volumetric flow, the centrifugal com-
at an elevated temperature. pressor experiences periodic backflow and surging.
Desuperheating suction gas at intermediate pressures where mul- Figure 1 shows an example system of curves representing the
tistage compressors balance is essential to reduce discharge temper- maximum refrigeration capacities for a brine chilling plant. The
atures of the upper-stage compressor. Desuperheating also helps example shows only one type of positive-displacement compressor
reduce oil carryover and reduces energy requirements. Subcooling using a water-cooled condenser in a single-stage system operating
improves the net refrigeration effect of the refrigerant supplied to the at a steady-state condition. The figure is a graphical method of
next-lower-temperature evaporator and reduces system energy re- expressing the first law of thermodynamics with an energy balance
quirements. The total heat is then rejected to a condenser. applied to a refrigeration system.
Subcoolers can be of shell-and-tube, shell-and-coil, welded- One set of nearly parallel curves (A) represents cooler capacity
plate, or tube-in-tube construction. Friction losses reduce the liquid at various brine temperatures versus saturated suction temperature
pressure that feeds refrigerant to an evaporator. Subcoolers are (a pressure condition) at the compressor, allowing for suction line
used to improve system efficiency and to prevent refrigerant liquid pressure drops. The (B) curves represent compressor capacities as
from flashing because of pressure loss caused by friction and the the saturated suction temperature varies and the saturated con-
vertical rise in lines. Refrigerant blends (zeotropes) can take ad- denser temperature (a pressure condition) varies. The (C) curves
vantage of temperature glide on the evaporator side with a direct- represent heat transferred to the condenser by the compressor. It is
expansion-in-tube serpentine or coil configuration. In this case, calculated by adding the heat input at the evaporator to the energy
temperature glide from the bubble point to the dew point promotes imparted to the refrigerant by the compressor. The (D) curves rep-
efficiency and lower surface requirements for the subcooler. A resent condenser performance at various saturated condenser tem-
flooded shell for the evaporating refrigerant requires use of only peratures as the inlet temperature of a fixed quantity of cooling
the higher dew-point temperature. water is varied.
Lubricant coolers remove friction heat and some of the super- The (E) curves represent the combined compressor and con-
heat of compression. Heat is usually removed by water, air, or a denser performance as a “condensing unit” at various saturated suc-
direct-expansion refrigerant. tion temperatures for various cooling water temperatures. These
Condensers that reject heat from the refrigeration system are curves were cross plotted from the (C) and (D) curves back to the set
available in many standard forms, such as water- or brine-cooled of brine cooler curves as indicated by the dashed construction lines
shell-and-tube, shell-and-coil, plate-and-frame, or tube-in-tube for the 27 and 33°C cooling water temperatures. Another set of con-
condensers; water cascading or sprayed over plate or coil serpentine struction lines (not shown) would be used for the 30°C cooling
models; and air-cooled, fin-coil condensers. Special heat pump con- water. The number of construction lines used can be increased as
densers are available in other forms, such as tube-in-earth and sub- necessary to adequately define curvature (usually no more than
merged tube bundle, or as serpentine and cylindrical coil condensers three per condensing-unit performance line).
that heat baths of boiling or single-phase fluids. The intersections of curves (A) and (E) represent the maximum
capacities for the entire system at those conditions. For example,
SELECTING DESIGN BALANCE POINTS these curves show that the system develops 532 kW when cooling
the brine to 7°C at 2.8°C (saturated) suction and using 27°C cooling
Refrigeration load at each designated evaporator pressure, refrig- water. At 33°C cooling water, capacity drops to 483 kW if the
erant properties, liquid refrigerant temperature feeding each evapo- required brine temperature is 6°C and the required saturated suction
rator, and evaporator design determine the required flow rate of temperature is 1.7°C. The corresponding saturated condensing tem-
refrigerant in a system. The additional flow rates of refrigerant that perature for 6°C brine with an accompanying suction temperature of
provide refrigerant liquid cooling, desuperheating, and compressor 2.8° and using 27°C water is graphically projected on the brine
lubricant cooling, where used, depend on the established liquid cooler line with a capacity of 532 kW to meet a newly constructed
refrigerant temperatures and intermediate pressures. 2.8°C saturated suction temperature line (parallel to the 1°C and
For a given refrigerant and flow rate, the suction line pressure 3°C lines). At this junction, draw a horizontal line to intersect the
drop, suction gas temperature, pressure ratio and displacement, and vertical saturated condensing temperature scale at 34.2°C. The con-
volumetric efficiency determine the required size and speed of rota- denser heat rejection is apparent from the (C) curves at a given bal-
tion for a positive displacement compressor. At low flow rates, par- ance point.
ticularly at very low temperatures and in long suction lines, heat The equation at the bottom of Figure 1 may be used to determine
gain through insulation can significantly raise the suction tempera- the shaft power required at the compressor for any given balance
ture. Also, at low flow rates, a large, warm compressor casing and point. A sixth set of curves could be drawn to indicate the power
suction plenum can further heat the refrigerant before it is com- requirement as a function of capacity versus saturated suction and
pressed. These heat gains increase the required displacement of a saturated condensing temperatures.
compressor. The compressor manufacturer must recommend the The same procedure can be repeated to calculate cascade system
superheating factors to apply. The final suction gas temperature performance. Rejected heat at the cascade condenser would be
from suction line heating is calculated by iteration. treated as the chiller load in making a cross plot of the upper-cycle,
Another concern is that more energy is required to compress high-temperature refrigeration system.
refrigerant to a given condenser pressure as the suction gas gains For cooling air at the evaporator(s) and for condenser heat rejec-
more superheat. This can be seen by examining a pressure-enthalpy tion to ambient air or evaporative condensers, use the same proce-
diagram for a given refrigerant such as R-22, which is shown in dures. Performance of coils and expansion devices such as
Figure 2 in Chapter 20 of the 2005 ASHRAE Handbook—Funda- thermostatic expansion valves may also be graphed, once the basic
mentals. As suction superheat increases along the horizontal axis, concept of heat and mechanical energy input equivalent combina-
the slopes of the constant entropy lines of compression decrease. tions is recognized. Chapter 1 of the 2005 ASHRAE Handbook—
This means that a greater enthalpy change must occur to produce a Fundamentals has further information.
given pressure rise. For a given flow, then, the power required for This method finds the natural balance points of compressors
compression is increased. With centrifugal compressors, pumping operating at their maximum capacities. For multiple-stage loads at
capacity is related to wheel diameter and speed, as well as to volu- several specific operating temperatures, the usual way of control-
metric flow and acoustic velocity of the refrigerant at the suction ling compressor capacities is with a suction pressure control and
Component Balancing in Refrigeration Systems 43.3

Fig. 1 Brine Chiller Balance Curve

Fig. 1 Brine Chiller Balance Curve

compressor capacity control device. This control accommodates of the system with designated pressures and temperatures, loads,
any mismatch in pumping capabilities of multistage compressors, enthalpies, flow rates, and energy requirements helps identify all
instead of allowing each compressor to find its natural balance important factors and components.
point. An overall energy and mass balance for the system is also essen-
Computer programs could be developed to determine balance tial to avoid mistakes. The overall system represented by the com-
points of complex systems. However, because applications, compo- plete flow diagram should be enclosed by a dotted line envelope.
nents, and piping arrangements are so diverse, many designers use Any energy inputs to or outputs from the system that directly affect
available capacity performance data from vendors and plot balance the heat content of the refrigerant itself should cross the dotted
points for chosen components. Individual computer programs may envelope line and must enter the energy balance equations. Accu-
be available for specific components, which speeds the process. rate estimates of the ambient heat gains through insulation and heat
losses from discharge lines where they are significant improve the
ENERGY AND MASS BALANCES comprehensiveness of the energy balance and accuracy of equip-
ment selections.
A systematic, point-to-point flow analysis of the system (includ- Cascade condenser loads and subcooler or desuperheating loads
ing piping) is essential in accounting for pressure drops and heat carried by a refrigerant are internal to the system and thus do not
gains, particularly in long suction lines. Air-cooled condensers, in enter into the overall energy balance. The total energy entering the
particular, can have large pressure drops, which must be included in system equals the total energy leaving the system. If calculations do
the analysis to estimate a realistic balance. Making a flow diagram not show an energy balance within reasonable tolerances for the
43.4 2006 ASHRAE Handbook—Refrigeration (SI)

accuracy of data used, then an omission occurred or a mistake was Gears and belt or chain drives have friction and windage power
made and should be corrected. losses that must be included in any meaningful analysis.
The dotted envelope technique can be applied to any section of Stack gas flows and enthalpies for engine or gas turbine exhausts
the system, but all energy transmissions must be included in the as well as air inputs and speeds must be included. In this case, per-
equations, including the enthalpies and mass flow rates of streams formance curves issued by the vendor must be heavily relied on to
that cross the dotted line. estimate the energy input to the system.
Calculating steam turbine performance requires measurements
SYSTEM PERFORMANCE of turbine speed, steam pressures and temperatures, and condensate
mass flow coupled with confidence that the vendor’s performance
Rarely are sufficient sensors and instrumentation devices avail- curves truly represent the current mechanical condition. Plant per-
able, nor are conditions proper at a given job site to allow calcula- sonnel normally have difficulty in obtaining operating data at spec-
tion of a comprehensive, accurate energy balance for an operating ified performance values.
system. Water-cooled condensers and oil coolers for heat rejection Heat rejection from air-cooled and evaporative condensers or
and the use of electric motor drives, where motor efficiency and coolers is extremely difficult to measure accurately because of
power factor curves are available, offer the best hope for estimat- changing ambient temperatures and the extent and scope of airflow
ing the actual performance of the individual components in a sys- measurements required. Often, one of the most important issues is
tem. Evaporator heat loads can be derived from the measured heat the wide variation or cycling of process flows, process tempera-
rejection and derived mechanical or measured electrical energy tures, and product refrigeration loads. Hot-gas false loading and
inputs. A comprehensive flow diagram assists in a field survey. compressor continuous capacity modulations complicate any
Various coolant flow detection devices are available for direct attempt to make a meaningful analysis.
measurement inside a pipe and for measurement from outside the Prediction and measurement of performance of systems using
pipe with variable degrees of accuracy. Sometimes flow rates may refrigerant blends (zeotropes) are especially challenging because of
be estimated by simply weighing or measuring an accumulation of temperature variations between bubble points and dew points.
coolant over a brief time interval. Nevertheless, ideal conditions of nearly steady-state loads and
Temperature and pressure measurement devices should be cali- flows with a minimum of cycling sometimes occur frequently
brated and be of sufficient accuracy. Calibrated digital scanning enough to permit a reasonable analysis. Computer-controlled sys-
devices for comprehensive simultaneous readings are best. Electri- tems can provide the necessary data for a more accurate system
cal power meters are not always available, so voltage and current at analysis. Several sets of nearly simultaneous data at all points over
each leg of a motor power connection must be measured. Voltage a short time enhance the accuracy of any calculation of performance
drops for long power leads must be calculated when the voltage of a given system. In all cases, properly purging condensers and
measurement points are far removed from the motor. Motor load eliminating excessive lubricant contamination of the refrigerant at
versus efficiency and power factor curves must be used to determine the evaporators are essential to determine system capabilities accu-
motor output to the system. rately.

Related Commercial Resources