CHAPTER 6

CONTROL OF MOISTURE AND OTHER CONTAMINANTS

IN REFRIGERANT SYSTEMS
Moisture .......................................................................................................................................... 6.1
Other Contaminants ........................................................................................................................ 6.6
System Cleanup Procedure After Hermetic Motor Burnout ........................................................... 6.8
Contaminant Control During Retrofit ............................................................................................. 6.9
Chiller Decontamination ............................................................................................................... 6.10

MOISTURE

hydrate. Ice forms during refrigerant evaporation when the relative

saturation of vapor reaches 100% at temperatures of 32F or below.
The separation of water as ice or liquid also is related to the solubility of water in a refrigerant. This solubility varies for different
refrigerants and with temperature (Table 1). Various investigators
have obtained different results on water solubility in R-134a and
R-123. The data presented here are the best available. The greater
the solubility of water in a refrigerant, the less the possibility that ice
or liquid water will separate in a refrigerating system. The solubility
of water in ammonia, carbon dioxide, and sulfur dioxide is so high
that ice or liquid water separation does not occur.
The concentration of water by mass at equilibrium is greater in
the gas phase than in the liquid phase of R-12 (Elsey and Flowers

OISTURE (water) is an important and universal contaminant

in refrigeration systems. The amount of moisture in a refrigerant system must be kept below an allowable maximum for satisfactory operation. Moisture must be removed from components
during manufacture and assembly to minimize the amount of moisture in the completed assembly. Any moisture that enters during
installation or servicing should be removed promptly.

Sources of Moisture
Moisture in a refrigerant system results from

Inadequate equipment drying in factories and service operations

Introduction during installation or service operations in the field
Low-side leaks, resulting in entrance of moisture-laden air
Leakage of water-cooled heat exchangers
Oxidation of some hydrocarbon lubricants that produce moisture
Wet lubricant, refrigerant, or desiccant
Moisture entering a nonhermetic refrigerant system through
hoses and seals

Table 1

Solubility of Water in Liquid Phase of

Certain Refrigerants

Solubility, ppm (by weight)

Temp.,
F
R-11 R-12 R-13 R-22 R-113 R-114 R-123 R-134a R-502

Drying equipment in the factory is discussed in Chapter 43.

Proper installation and service procedures as given in ASHRAE
Standard 147 minimize the second, third, and fourth sources. Lubricants are discussed in Chapter 7. If purchased refrigerants and lubricants meet specifications and are properly handled, the moisture
content generally remains satisfactory. See the section on Electrical
Insulation under Compatibility of Materials in Chapter 5 and the
section on Motor Burnouts in this chapter.

Effects of Moisture
Excess moisture in a refrigerating system can cause one or all of
the following undesirable effects:

Ice formation in expansion valves, capillary tubes, or evaporators

Corrosion of metals
Copper plating
Chemical damage to motor insulation in hermetic compressors or
other system materials
Hydrolysis of lubricants and other materials
Sludge formation

160
150
140
130
120
110

460
400
340
290
240
200

700
560
440
350
270
210

4100
3600
3150
2750
2400
2100

460
400
344
290
240
200

450
380
320
270
220
180

2600
2300
2000
1800
1600
1400

4200
3600
3200
2800
2400
2000

1780
1580
1400
1220
1080
930

100
90
80
70
60

168
140
113
90
70

165
128
98
76
58

44

1800
1580
1350
1140
970

168
140
113
90
70

148
120
95
74
57

1200
1000
900
770
660

1800
1500
1300
1100
880

810
690
580
490
400

50
40
30
20
10

55
44
34
26
20

44
32
23.3
16.6
11.8

26

14

830
690
573
472
384

55
44
34
26
20

44
33
25
18
13

560
470
400
330
270

730
600
490
390
320

335
278
225
180
146

0
10
20
30
40

15
11
8
6
4

8.3
5.7
3.8
2.5
1.7

308
244
195
152
120

15
11
8
6

10
7
5
3
2

220
180
140
110
90

250
200
150
120
89

115
90
69
53
40

1.1
0.7
0.4
0.3
0.1
0.1

91
68
50
37
27
19

70
53
40
30
22
16

66
49
35
25
18
12

30
22
16
11
8
5

50
60
70
80
90
100

Ice or solid hydrate separates from refrigerants if the water concentration is high enough and the temperature low enough. Solid
hydrate, a complex molecule of refrigerant and water, can form at
temperatures higher than those required to separate ice. Liquid water
forms at temperatures above those required to separate ice or solid

3
2
1
0.8
0.5
0.3

1.5
1
0.6
0.4
0.2
0.1

Data on R-134a adapted from Thrasher et al. (1993) and Allied-Signal Corporation.
Data on R-123 adapted from Thrasher et al. (1993) and E.I. DuPont de Nemours &
Company. Remaining data adapted from E.I. DuPont de Nemours & Company and
Allied-Signal Corporation.

The preparation of this chapter is assigned to TC 3.3, Refrigerant Contaminant Control.

6.1

6.2

2006 ASHRAE HandbookRefrigeration

3.8 ppm 15.3 = 58 ppm

Table 2 Distribution of Water Between Vapor and

Liquid Phases of Certain Refrigerants
Temperature,
F

Water in Vapor/Water in Liquid, mass %/mass %

R-11 R-12

R-22

100
90
80
70

30.1

34.8

5.5
6.1
6.3
7.5

0.400
0.397
0.405
0.404

R-114 R-123 R-134a R-502

13.3
14.4
16.0
17.6

4.25
4.57
4.53
4.69

0.542
0.571
0.573
0.585

0.63
0.64
0.65
0.66

60
50
40
30

43.1

50.5

8.2
9.0
9.9
11.2

0.401
0.391
0.390
0.378

19.7
21.7
24.7
26.9

4.82
4.97
5.16
5.17

0.625
0.637
0.650
0.654

0.66
0.66
0.66
0.65

20
0
20
40

57.9
62.4
71.0

11.9
13.1
15.3
17.1

0.351
0.301
0.251
0.203

29.5
32.1
37.4
52.2

5.09
4.91
4.78
4.43

0.639
0.584
0.543
0.484

0.61
0.53
0.47
0.40

Data adapted from E.I. DuPont de Nemours & Company, Inc.

1949). The opposite is true for Refrigerants 22 and 502. The ratio
of mass concentrations differs for each refrigerant; it also varies
with temperature. Table 2 shows the distribution ratios of water in
the vapor phase to water in the liquid phase for common refrigerants. It can be used to calculate the equilibrium water concentration
of the liquid-phase refrigerant if the gas phase concentration is
known, and vice versa. The water content in the vapor phase is
determined by
Pw
W = -------------0( P w)

( d w)
------------0
( d R)

(1)

where
W=
Pw =
(P w)0 =
(dw)0 =
(dR)0 =

mass water/mass refrigerant

partial pressure of water vapor
partial pressure of water vapor at saturation
density of water vapor at saturation
density of refrigerant vapor at saturation

3.8 0.587 = 2.2 ppm

58.0 0.413 = 24.0 ppm
26.2 ppm
Maintaining moisture levels below critical value keeps free water
from the low side of the system.
The previous analysis can be applied to all refrigerants and applications. An R-22 system with 110F liquid and 20F evaporating
temperatures reaches saturation when the moisture circulating is
139 ppm. Note that this value is less than the liquid solubility,
195 ppm at 20F.
Excess moisture causes paper or polyester motor insulation to
become brittle, which can cause premature motor failure. However,
not all motor insulations are affected adversely by moisture. The
amount of water in a refrigerant system must be small enough to avoid
ice separation, corrosion, and insulation breakdown.
Polyol ester lubricants (POEs), which are used largely with hydrofluorocarbons (HFCs), absorb substantially more moisture than
do mineral oils, and do so very rapidly on exposure to the atmosphere. Once present, the moisture is difficult to remove. Hydrolysis
of POEs can lead to formation of acids and alcohols that, in turn, can
negatively affect system durability and performance (Griffith 1993).
For these reasons, POEs should not be exposed to ambient air except
for very brief periods required for compressor installation. Also, adequate driers are particularly important elements for equipment
containing POEs.
Exact experimental data on the maximum permissible moisture
level in refrigerant systems are not known because so many factors
are involved.

Drying Methods

Freezing at expansion valves or capillary tubes can occur when

excessive moisture is present in a refrigerating system. Formation of
ice or hydrate in evaporators can partially insulate the evaporator
and reduce efficiency or cause system failure. Excess moisture can
cause corrosion and enhance copper plating (Walker et al. 1962).
Other factors affecting copper plating are discussed in Chapter 5.
The moisture required for freeze-up is a function of the amount
of refrigerant vapor formed during expansion and the distribution of
water between the liquid and gas phases downstream of the expansion device. For example, in an R-12 system with a 110F liquid
temperature and a 20F evaporator temperature, refrigerant after
expansion is 41.3% vapor and 58.7% liquid (by mass). The percentage of vapor formed is determined by
h L( liquid ) h L( eva p )
% Vapor = 100 -----------------------------------------------h fg( evap )

When the vapor contains more than the saturation quantity

(100% rh), free water will be present as a third phase. If the temperature is below 32F, ice will form. Using the saturated moisture values and the liquid-vapor ratios, the critical water content of the
circulating refrigerant can be calculated as

(2)

where
hL(liquid) = saturated liquid enthalpy for refrigerant at liquid temperature
hL(evap) = saturated liquid enthalpy for refrigerant at evaporating
temperature
hfg(evap) = latent heat of vaporization of refrigerant at evaporating
temperature

Table 1 lists the saturated water content of the R-12 liquid phase
at 20F as 3.8 ppm. Table 2 is used to determine the saturated vapor
phase water content as

Equipment in the field is dried by decontamination, evacuation,

and driers. Before opening equipment for service, refrigerant must
be isolated or recovered into an external storage container (see
Chapter 8). After installation or service, noncondensable gases (air)
should be removed with a vacuum pump connected preferably to
both suction and discharge service ports. The absolute pressure
should be reduced to 1 mm of mercury or less, which is below the
vapor pressure of water at ambient temperature. External or internal
heat may be required to vaporize water in the system. Take care not
to overheat the equipment. Even with these procedures, small
amounts of moisture trapped under a lubricant film, adsorbed by the
motor windings, or located far from the vacuum pump are difficult
to remove. Evacuation will not remove any significant amount of
water from polyol ester lubricants used in HFC systems.
It is good practice to install a drier. Larger systems frequently use
a drier with a replaceable core. The core may need to be changed
several times before the proper degree of dryness is obtained. A
moisture indicator in the liquid line can indicate when the system
has been dried satisfactorily.
Special techniques are required to remove free water in a refrigeration or air-conditioning system from a burst tube or water chiller
leak. Refrigerant should be transferred to a pumpdown receiver or
recovered in a separate storage tank. Parts of the system may have to
be disassembled and the water drained from system low points. In
some large systems, the semihermetic or open-drive compressor
may need to be cleaned by disassembling and hand-wiping the
various parts. Decontamination work should be performed before
reinstallation of compressors, particularly hermetic units. After

Control of Moisture and Other Contaminants in Refrigerant Systems

reassembly, the compressor should be dried further by passing dry
nitrogen through the system and by heating and evacuation. Using
internal heat, by circulating warmed water on the water side of
water-cooled equipment, is preferred. Drying may take an extended
period and require frequent changes of the vacuum pump lubricant.
Liquid-line driers should be replaced and temporary suction line
driers installed. During initial operation, driers need to be changed
often. Decontamination procedures use large temporary driers.
Properly performed decontamination eliminates the need for frequent on-board liquid-line drier changes.
If refrigerant in the pumpdown receiver is to be reused, it must be
thoroughly dried before being reintroduced into the system. One
method begins by drawing a liquid refrigerant sample and recording
the ambient temperature. If chemical analysis of the sample by a
qualified laboratory reveals a moisture content at or near the water
solubility in Table 1 at the recorded temperature, then free water is
probably present. In that case, a recovery unit with a suction filterdrier and/or a moisture/lubricant trap must be used to transfer the
bulk of the refrigerant from the receiver liquid port to a separate
tank. When the free water reaches the tank liquid port, most of the
remaining refrigerant can be recovered through the receiver vapor
port. The water can then be drained from the pumpdown receiver.

Moisture Indicators
Moisture-sensitive elements that change color according to
moisture content can gage the systems moisture level; the color
changes at a low enough level to be safe. Manufacturers instructions must be followed because the color change point is also
affected by liquid-line temperature and the refrigerant used.

Moisture Measurement
Techniques for measuring the amount of moisture in a compressor, or in an entire system, are discussed in Chapter 45. The following methods are used to measure the moisture content of various
halocarbon refrigerants. The moisture content to be measured is
generally in the parts-per-million range, and the procedures require
special laboratory equipment and techniques.
The Karl Fischer method is suitable for measuring the moisture
content of a refrigerant, even if it contains mineral oil. Although different firms have slightly different ways of performing this test and
get somewhat varying results, the method remains the common
industry practice for determining moisture content in refrigerants.
The refrigerant sample is bubbled through predried methyl alcohol
in a special sealed glass flask; any water present remains with the
alcohol. In volumetric titration, Karl Fischer reagent is added, and
the solution is immediately titrated to a dead stop electrometric
end point. The reagent reacts with any moisture present so that the
amount of water in the sample can be calculated from a previous calibration of the Karl Fischer reagent.
In coulometric titration (ARI Standard 700c), water is titrated
with iodine that is generated electrochemically. The instrument
measures the quantity of electric charge used to produce the iodine
and titrate the water and calculates the amount of water present.
These titration methods, considered among the most accurate,
are also suitable for measuring the moisture content of unused lubricant or other liquids. Special instruments designed for this particular analysis are available from laboratory supply companies.
Haagen-Smit et al. (1970) describe improvements in the equipment
and technique that significantly reduce analysis time.
The gravimetric method for measuring moisture content of
refrigerants is described in ASHRAE Standards 35 and 63.1. It is
not widely used in the industry. In this method, a measured
amount of refrigerant vapor is passed through two tubes in series,
each containing phosphorous pentoxide (P2O5). Moisture present
in the refrigerant reacts chemically with the P2O5 and appears as
an increase in mass in the first tube. The second tube is used as a
tare. This method is satisfactory when the refrigerant is pure, but

6.3

the presence of lubricant produces inaccurate results, because the

lubricant is weighed as moisture. Approximately 200 g of refrigerant is required for accurate results. Because the refrigerant must
pass slowly through the tube, analysis requires many hours.
DeGeiso and Stalzer (1969) discuss the electrolytic moisture
analyzer, which is suitable for high-purity refrigerants. Other electronic hygrometers are available that sense moisture by the adsorption of water on an anodized aluminum strip with a gold foil overlay
(Dunne and Clancy 1984). Calibration is critical to obtain maximum
accuracy. These hygrometers give a continuous moisture reading
and respond rapidly enough to monitor changes. Data showing drydown rates can be gathered with these instruments (Cohen 1994).
Brisken (1955) used this method in a study of moisture migration in
hermetic equipment.
Thrasher et al. (1993) used nuclear magnetic resonance spectroscopy to determine the moisture solubilities in R-134a and R-123.
Another method, infrared spectroscopy, is used for moisture analysis, but requires a large sample for precise results and is subject to
interference if lubricant is present in the refrigerant.

Desiccants
Desiccants used in refrigeration systems adsorb or react chemically with the moisture contained in a liquid or gaseous refrigerant/
lubricant mixture. Solid desiccants, used widely as dehydrating
agents in refrigerant systems, remove moisture from both new and
field-installed equipment. The desiccant is contained in a device
called a drier (also spelled dryer) or filter-drier and can be installed
in either the liquid or the suction line of a refrigeration system.
Desiccants must remove most of the moisture and not react unfavorably with any other materials in the system. Activated alumina,
silica gel, and molecular sieves are the most widely used desiccants
acceptable for refrigerant drying. Water is physically adsorbed on
the internal surfaces of these highly porous desiccant materials.
Activated alumina and silica gel have a wide range of pore sizes,
which are large enough to adsorb refrigerant, lubricant, additives,
and water molecules. Pore sizes of molecular sieves, however, are
uniform, with an aperture of approximately 0.3 nm for a type 3A
molecular sieve or 0.4 nm for a type 4A molecular sieve. The uniform openings exclude lubricant molecules from the adsorption
surfaces. Molecular sieves can be selected to exclude refrigerant
molecules, as well. This property gives the molecular sieve the advantage of increasing water capacity and improving chemical compatibility between refrigerant and desiccant (Cohen 1993, 1994;
Cohen and Blackwell 1995). The drier or desiccant manufacturer
can provide information about which desiccant adsorbs or excludes
a particular refrigerant.
Drier manufacturers offer combinations of desiccants that can be
used in a single drier and may have advantages over a single desiccant because they can adsorb a greater variety of refrigeration contaminants. Two combinations are activated alumina with molecular
sieves and silica gel with molecular sieves. Activated carbon is also
used in some mixtures.
Desiccants are available in granular, bead, and block forms. Solid
core desiccants, or block forms, consist of desiccant beads, granules,
or both held together by a binder (Walker 1963). The binder is
usually a nondesiccant material. Suitable filtration, adequate contact
between desiccant and refrigerant, and low pressure drop are
obtained by properly sizing the desiccant particles used to make up
the core, and by the proper geometry of the core with respect to the
flowing refrigerant. Beaded molecular sieve desiccants have higher
water capacity per unit mass than solid-core desiccants. The composition and form of the desiccant is varied by drier manufacturers to
achieve the desired properties.
Desiccants that take up water by chemical reaction are not recommended. Calcium chloride reacts with water to form a corrosive liquid.
Barium oxide is known to cause explosions. Magnesium perchlorate
and barium perchlorate are powerful oxidizing agents, which are

6.4

2006 ASHRAE HandbookRefrigeration

Table 3

Reactivation of Desiccants

Desiccant
Activated alumina
Silica gel
Molecular sieves

Temperature, F

Fig. 1 Moisture Equilibrium Curves for R-12 and

Three Common Desiccants at 75F

400 to 600
350 to 600
500 to 660

potential explosion hazards in the presence of lubricant. Phosphorous

pentoxide is an excellent desiccant, but its fine powdery form makes it
difficult to handle and produces a high resistance to gas and liquid
flow. A mixture of calcium oxide and sodium hydroxide, which has
limited use as an acid scavenger, should not be used as a desiccant.
Desiccants readily adsorb moisture and must be protected
against it until ready for use. If a desiccant has picked up moisture,
it can be reactivated under laboratory conditions by heating for
about 4 h at a suitable temperature, preferably with a dry-air purge
or in a vacuum oven (Table 3). Only adsorbed water is driven off at
the temperatures listed, and the desiccant is returned to its initial
activated state. Avoid repeated reactivation and excessive temperatures during reactivation, which may damage the desiccant. Desiccant in a refrigerating equipment drier should not be reactivated for
reuse, because of lubricant and other contaminants in the drier as
well as possible damage caused by overheating the drier shell.
Equilibrium Conditions of Desiccants. Desiccants in refrigeration and air-conditioning systems function on the equilibrium principle. If an activated desiccant contacts a moisture-laden refrigerant,
the water is adsorbed from the refrigerant/water mixture onto the desiccant surface until the vapor pressures of the adsorbed water (i.e., at
the desiccant surface) and the water remaining in the refrigerant are
equal. Conversely, if the vapor pressure of water on the desiccant surface is higher than that in the refrigerant, water is released into the
refrigerant/water mixture, and equilibrium is reestablished.
Adsorbent desiccants function by holding (adsorbing) moisture
on their internal surfaces. The amount of water adsorbed from a refrigerant by an adsorbent at equilibrium is influenced by (1) pore
volume, pore size, and surface characteristics of the adsorbent;
(2) temperature and moisture content of the refrigerant; and (3) solubility of water in the refrigerant.
Figures 1 to 3 are equilibrium curves (known as adsorption isotherms) for various adsorbent desiccants with R-12 and R-22.
These curves are representative of commercially available materials. The adsorption isotherms are based on the technique developed
by Gully et al. (1954), as modified by ASHRAE Standard 35.
ASHRAE Standards 35 and 63.1 define the moisture content of the
refrigerant as equilibrium point dryness (EPD), and the moisture
held by the desiccant as water capacity. The curves show that for any
specified amount of water in a particular refrigerant, the desiccant
holds a corresponding specific quantity of water.
Figures 1 and 2 show moisture equilibrium curves for three common adsorbent desiccants in drying R-12 and R-22 at 75F. As
shown, desiccant capacity can vary widely for different refrigerants
when the same EPD is required. Generally, a refrigerant in which
moisture is more soluble requires more desiccant for adequate drying than one that has less solubility.
Figure 3 shows the effect of temperature on moisture equilibrium capacities of activated alumina and R-12. Much higher water
capacities are obtained at lower temperatures, demonstrating the
advantage of locating alumina driers at relatively cool spots in the
system. The effect of temperature on molecular sieves water
capacity is much smaller. ARI Standard 710 requires determining
the water capacity for R-12 at an EPD of 15 ppm, and for R-22 at
60 ppm. Each determination must be made at 75F (see Figures 1
and 2) and 125F.
Figure 4 shows water capacity of a molecular sieve in liquid
R-134a at 125F. These data were obtained using the Karl Fischer
method similar to that described in Dunne and Clancy (1984).
Cavestri and Schafer (1999) determined water capacities for three

LIVE GRAPH

Click here to view

Fig. 1

Moisture Equilibrium Curves for R-12 and Three

Common Desiccants at 75F

Fig. 2 Moisture Equilibrium Curves for R-22 and

Three Common Desiccants at 75F

LIVE GRAPH

Click here to view

Fig. 2

Moisture Equilibrium Curves for R-22 and Three

Common Desiccants at 75F

common desiccants in R-134a when POE lubricant was added to

the refrigerant. Figures 5 and 6 show water capacity for type 3A
molecular sieves, activated alumina beads, and bonded activated
alumina cores in R-134a and 2% POE lubricant at 75F and 125F.
Although the figures show that molecular sieves have greater
water capacities than activated alumina or silica gel at the indicated EPD, all three desiccants are suitable if sufficient quantities
are used. Cost, operating temperature, other contaminants present,
and equilibrium capacity at the desired EPD must be considered
when choosing a desiccant for refrigerant drying. Consult the desiccant manufacturer for information and equilibrium curves for
specific desiccant/refrigerant systems.

Control of Moisture and Other Contaminants in Refrigerant Systems

Fig. 3 Moisture Equilibrium Curves for Activated Alumina
at Various Temperatures in R-12

LIVE GRAPH

Click here to view

Fig. 3

Moisture Equilibrium Curves for Activated Alumina

at Various Temperatures in R-12

Fig. 4 Moisture Equilibrium Curve for Molecular Sieve

in R-134a at 125F

6.5

Fig. 5 Moisture Equilibrium Curves for Three Common Desiccants in R-134a and 2% POE Lubricant at 75F

LIVE GRAPH

Click here to view

Fig. 5 Moisture Equilibrium Curves for Three Common

Desiccants in R-134a and 2% POE Lubricant at 75F
Fig. 6 Moisture Equilibrium Curves for Three Common Desiccants in R-134a and 2% POE Lubricant at 125F

LIVE GRAPH

Click here to view

Fig. 4

Moisture Equilibrium Curve for Molecular Sieve

in R-134a at 125F
(Courtesy UOP, Reprinted with permission.)

Activated carbon technically is not a desiccant, but it is often

used in filter-driers to scavenge waxes and insoluble resins. The
other common desiccants do not remove these contaminants, which
can plug expansion devices and reduce system capacity and efficiency. Activated carbon is typically incorporated into bound desiccant blocks along with molecular sieve and activated alumina.

Desiccant Applications
In addition to removing water, desiccants may adsorb or react
with acids, dyes, chemical additives, and refrigerant lubricant reaction products.
Acids. Generally, acids can harm refrigerant systems. The
amount of acid a refrigerant system tolerates depends in part on the
size, mechanical design and construction of the system, type of
motor insulation, type of acid, and amount of water in the system.
Desiccants acid removal capacity is difficult to determine because the environment is complex. Hoffman and Lange (1962) and
Mays (1962) showed that desiccants remove acids from refrigerants
and lubricants by adsorption and/or chemical reaction. Hoffman and
Lange also showed that the loading of water on the desiccant, type
of desiccant, and type of acid play major roles in a desiccants ability to remove acids from refrigerant systems. In addition, acids
formed in these systems can be inorganic, such as HCl and HF, or a
mixture of organic acids. All of these factors must be considered to
establish acid capacities of desiccants.

LIVE GRAPH

Click here to view

Fig. 6 Moisture Equilibrium Curves for Three Common

Desiccants in R-134a and 2% POE Lubricant at 125F
Cavestri and Schooley (1998) determined the inorganic acid
capacity of desiccants. Both molecular sieve and alumina desiccants
remove inorganic acids such as HCl and HF from refrigerant systems. Molecular sieves remove these acids through irreversible
chemisorption: the acids form their respective salts with the molecular sieves sodium and potassium cations. Alumina removes such
acids principally by reversible physical adsorption.
Colors. Colored materials frequently are adsorbed by activated
alumina and silica gel and occasionally by calcium sulfate and
molecular sieves. Leak detector dyes may lose their effectiveness in
systems containing desiccants. The interaction of the dye and
drier should be evaluated before putting a dye in the system.
Lubricant Deterioration Products. Lubricants can react chemically to produce substances that are adsorbed by desiccants. Some
of these are hydrophobic and, when adsorbed by the desiccant, may
reduce the rate at which it can adsorb liquid water. However, the rate
and capacity of the desiccant to remove water dissolved in the
refrigerant are not significantly impaired (Walker et al. 1955).
Often, reaction products are sludges or powders that can be filtered
out mechanically by the drier.
Chemicals. Refrigerants that can be adsorbed by desiccants
cause the drier temperature to rise considerably when the refrigerant
is first admitted. This temperature rise is not the result of moisture in
the refrigerant, but the adsorption heat of the refrigerant. Lubricant

6.6
additives may be adsorbed by silica gel and activated alumina.
Because of small pore size, molecular sieves generally do not adsorb
additives or lubricant.

Driers
A drier is a device containing a desiccant. It collects and holds
moisture, but also acts as a filter and adsorber of acids and other contaminants.
To prevent moisture from freezing in the expansion valve or capillary tube, a drier is installed in the liquid line close to these devices.
Hot locations should be avoided. Driers can function on the lowpressure side of expansion devices, but this is not the preferred location (Jones 1969).
Moisture is reduced as liquid refrigerant passes through a drier.
However, Krause et al. (1960) showed that considerable time is
required to reach moisture equilibrium in a refrigeration unit. The
moisture is usually distributed throughout the entire system, and time
is required for the circulating refrigerant/lubricant mixture to carry
the moisture to the drier. Cohen (1994) and Cohen and Dunne (1987)
discuss the kinetics of drying refrigerants in circulating systems.
Loose-filled driers should be mounted vertically, with downward
refrigerant flow. In this configuration, both gravity and drag forces
act in the downward direction on the beads. Settling of the beads
creates a void space at the top, which is not a problem.
Vertical orientation with upward flow, where gravity and drag act
in opposite directions, should be avoided because the flow will
likely fluidize the desiccant beads, causing the beads to move
against each another. This promotes attrition or abrasion of the
beads, producing fine particles that can contaminate the system.
Settling creates a void space between the retention screens, promoting fluidization.
Horizontal mounting should also be avoided with a loose-filled
drier because bead settling creates a void space that promotes fluidization, and may also produce a channel around the beads that
reduces drying effectiveness.
Driers are also used effectively to clean systems severely contaminated by hermetic motor burnouts and mechanical failures (see
the section on System Cleanup Procedure after Hermetic Motor
Burnout).

Drier Selection
The drier manufacturers selection chart lists the amount of desiccants, flow capacity, filter area, water capacity, and a specific recommendation on the type and refrigeration capacity of the drier for
various applications.
The equipment manufacturer must consider the following factors
when selecting a drier:
1. The desiccant is the heart of the drier and its selection is most
important. The section on Desiccants has further information.
2. The driers water capacity is measured as described in ARI Standard 710. Reference points are set arbitrarily to prevent confusion
arising from determinations made at other points. The specific
refrigerant, amount of desiccant, and effect of temperature are all
considered in the statement of water capacity.
3. The liquid-line flow capacity is listed at 1 psi pressure drop
across the drier by the official procedures of ARI Standard 710
and ANSI/ASHRAE Standard 63.1. Rosen et al. (1965) described a closed-loop method for evaluating filtration and flow
characteristics of liquid-line refrigerant driers. The flow capacity
of suction-line filters and filter-driers is determined according
to ARI Standard 730 and ANSI/ASHRAE Standard 78. ARI
Standard 730 gives recommended pressure drops for selecting
suction-line filter-driers for permanent and temporary installations. Flow capacity may be reduced quickly when critical quantities of solids and semisolids are filtered out by the drier.

2006 ASHRAE HandbookRefrigeration

Whenever flow capacity drops below the machines requirements, the drier should be replaced.
4. Although limits for particle size vary with refrigerant system size
and design, and with the geometry and hardness of the particles,
manufacturers publish filtration capabilities for comparison.

Testing and Rating

Desiccants and driers are tested according to the procedures of
ASHRAE Standards 35 and 63. Driers are rated under ARI Standard 710. Minimum standards for listing of refrigerant driers can be
found in UL Standard 207. ANSI/ASHRAE Standard 63.2 specifies
a test method for filtration testing of filter-driers. No ARI Standard
has been developed to give rating conditions for publication of filtration capacity.

OTHER CONTAMINANTS
Refrigerant filter-driers are the principal devices used to remove
contaminants from refrigeration systems. The filter-drier is not a
substitute for good workmanship or design, but a maintenance tool
necessary for continued and proper system performance. Contaminants removed by filter-driers include moisture, acids, hydrocarbons with a high molecular weight, oil decomposition products, and
insoluble material, such as metallic particles and copper oxide.

Metallic Contaminants and Dirt

Small contaminant particles frequently left in refrigerating systems during manufacture or servicing include chips of copper, steel,
or aluminum; copper or iron oxide; copper or iron chloride; welding
scale; brazing or soldering flux; sand; and other dirt. Some of these
contaminants, such as copper chloride, develop from normal wear
or chemical breakdown during system operation. Solid contaminants vary widely in size, shape, and density. Solid contaminants
create problems by
Scoring cylinder walls and bearings
Lodging in the motor insulation of a hermetic system, where they
act as conductors between individual motor windings or abrade
the wire coating when flexing of the windings occurs
Depositing on terminal blocks and serving as a conductor
Plugging expansion valve screen or capillary tubing
Depositing on suction or discharge valve seats, significantly
reducing compressor efficiency
Plugging oil holes in compressor parts, leading to improper lubrication
Increasing the rate of chemical breakdown [e.g., at elevated temperatures, R-22 decomposes more readily when in contact with
iron powder, iron oxide, or copper oxide (Norton 1957)]
Plugging driers
Liquid-line filter-driers, suction filters, and strainers isolate contaminants from the compressor and expansion valve. Filters minimize return of particulate matter to the compressor and expansion
valve, but the capacity of permanently installed liquid and/or suction
filters must accommodate this particulate matter without causing
excessive, energy-consuming pressure losses. Equipment manufacturers should consider the following procedures to ensure proper
operation during the design life:
1. Develop cleanliness specifications that include a reasonable
value for maximum residual matter. Some manufacturers specify
allowable quantities in terms of internal surface area. ASTM
Standard B280 allows a maximum of 0.0035 g of contaminants
per square foot of internal surface.
2. Multiply the factory contaminant level by a factor of five to allow
for solid contaminants added during installation. This factor
depends on the type of system and the previous experience of the
installers, among other considerations.

Control of Moisture and Other Contaminants in Refrigerant Systems

3. Determine maximum pressure drop to be incurred by the suction
or liquid filter when loaded with the quantity of solid matter calculated in Step 2.
4. Conduct pressure drop tests according to ASHRAE Standard
63.2.
5. Select driers for each system according to its capacity requirements and test data. In addition to contaminant removal capacity,
tests can evaluate filter efficiency, maximum escaped particle
size, and average escaped particle size.
Very small particles passing through filters tend to accumulate in
the crankcase. Most compressors tolerate a small quantity of these
particles without allowing them into the oil pump inlet, where they
can damage running surfaces.

Organic Contaminants: Sludge, Wax, and Tars

Organic contaminants in a refrigerating system with a mineral oil
lubricant can appear when organic materials such as oil, insulation,
varnish, gaskets, and adhesives decompose. As opposed to inorganic
contaminants, these materials are mostly carbon, hydrogen, and oxygen. Organic materials may be partially soluble in the refrigerant/
lubricant mixture or may become so when heated. They then circulate in the refrigerating system and can plug small orifices. Organic
contaminants in a refrigerating system using a synthetic polyol ester
lubricant may also generate sludge. The following contaminants
should be avoided:
Paraffin (typically found in mineral oil lubricants)
Silicone (found in some machine lubricants)
Phthalate (found in some machine lubricants)
Whether mineral oil or synthetic lubricants are used, some
organic contaminants remain in a new refrigerating system during manufacture or assembly. For example, excessive brazing
paste introduces a waxlike contaminant into the refrigerant
stream. Certain cutting lubricants, corrosion inhibitors, or drawing compounds frequently contain paraffin-based compounds.
These lubricants can leave a layer of paraffin on a component that
may be removed by the refrigerant/lubricant combination and
generate insoluble material in the refrigerant stream. Organic
contamination also results during the normal method of fabricating return bends. The die used during forming is lubricated with
these organic materials, and afterwards the return bend is brazed
to the tubes to form the evaporator and/or condenser. During
brazing, residual lubricant inside the tubing and bends can be
baked to a resinous deposit.
If organic materials are handled improperly, certain contaminants remain. Resins used in varnishes, wire coating, or casting sealers may not be cured properly and can dissolve in the refrigerant/
lubricant mixture. Solvents used in washing stators may be adsorbed by the wire film and later, during compressor operation,
carry chemically reactive organic extractables. Chips of varnish, insulation, or fibers can detach and circulate in the system. Portions of
improperly selected or cured rubber parts or gaskets can dissolve in
the refrigerant.
Refrigeration-grade mineral oil decomposes under adverse
conditions to form a resinous liquid or a solid frequently found on
refrigeration filter-driers. These mineral oils decompose noticeably when exposed for as little as 2 h to temperatures as low as
250F in an atmosphere of air or oxygen. The compressor manufacturer should perform all high-temperature dehydrating operations on the machines before adding the lubricant charge. In
addition, equipment manufacturers should not expose compressors to processes requiring high temperatures unless the compressors contain refrigerant or inert gas.
The result of organic contamination is frequently noticed at the
expansion device. Materials dissolved in the refrigerant/lubricant
mixture, under liquid line conditions, may precipitate at the lower

6.7

temperature in the expansion device, resulting in restricted or

plugged capillary tubes or sticky expansion valves. A few milligrams of these contaminants can render a system inoperative.
These materials have physical properties that range from a fluffy
powder to a solid resin entraining inorganic debris. If the contaminant is dissolved in the refrigerant/lubricant mixture in the liquid
line, it will not be removed by a filter-drier.
Chemical identification of these organic contaminants is very
difficult. Infrared spectroscopy can characterize the type of organic
groups present in contaminants. Materials found in actual systems
vary from waxlike aliphatic hydrocarbons to resinlike materials
containing double bonds, carbonyl groups, and carboxyl groups. In
some cases, organic compounds of copper and/or iron have been
identified.
These contaminants can be eliminated by carefully selecting
materials and strictly controlling cleanliness during manufacture
and assembly of the components as well as the final system.
Because heat degrades most organic materials and enhances chemical reactions, operating conditions with excessively high discharge
or bearing surface temperatures must be avoided to prevent formation of degradation products.

Residual Cleaning Agents

Mineral Oil Systems. Solvents used to clean compressor parts
are likely contaminants if left in refrigerating equipment. Solvents
in this category are considered pure liquids without additives. If
additives are present, they are reactive materials and should not be
in a refrigerating system. Some solvents are relatively harmless to
the chemical stability of the refrigerating system, whereas others
initiate or accelerate degradation reactions. For example, the common mineral spirits solvents are considered harmless. Other common compounds react rapidly with hydrocarbon lubricants (Elsey
et al. 1952).
Polyol Ester Lubricated Systems. Typical solvents used in
cleaning mineral oil systems are not compatible with polyol ester
lubricants. Several chemicals must be avoided to reduce or eliminate possible contamination and sludge generation. In addition to
paraffin, silicone, and phthalate contaminants, a small amount of the
following contaminants can cause system failure:
Chlorides (typically found in chlorinated solvents)
Acid or alkali (found in some water-based cleaning fluids)
Water (component of water-based cleaning fluids)

Noncondensable Gases
Gases, other than the refrigerant, are another contaminant frequently found in refrigerating systems. These gases result (1) from
incomplete evacuation, (2) when functional materials release sorbed
gases or decompose to form gases at an elevated temperature during
system operation, (3) through low-side leaks, and (4) from chemical
reactions during system operation. Chemically reactive gases, such
as hydrogen chloride, attack other components, and, in extreme
cases, the refrigerating unit fails.
Chemically inert gases, which do not liquefy in the condenser,
reduce cooling efficiency. The quantity of inert, noncondensable
gas that is harmful depends on the design and size of the refrigerating unit and on the nature of the refrigerant. Its presence contributes
to higher-than-normal head pressures and resultant higher discharge
temperatures, which speed up undesirable chemical reactions.
Gases found in hermetic refrigeration units include nitrogen,
oxygen, carbon dioxide, carbon monoxide, methane, and hydrogen. The first three gases originate from incomplete air evacuation
or a low-side leak. Carbon dioxide and carbon monoxide usually
form when organic insulation is overheated. Hydrogen has been
detected when a compressor is experiencing serious bearing wear.
These gases are also found where a significant refrigerant/lubricant

6.8

2006 ASHRAE HandbookRefrigeration

reaction has occurred. Only trace amounts of these gases are

present in well-designed, properly functioning equipment.
Doderer and Spauschus (1966), Gustafsson (1977), and Spauschus and Olsen (1959) developed sampling and analytical techniques for establishing the quantities of contaminant gases present
in refrigerating systems. Kvalnes (1965), Parmelee (1965), and
Spauschus and Doderer (1961, 1964) applied gas analysis techniques to sealed tube tests to yield information on stability limitations of refrigerants, in conjunction with other materials used in
hermetic systems.

Motor Burnouts
Motor burnout is the final result of hermetic motor insulation
failure. During burnout, high temperatures and arc discharges can
severely deteriorate the insulation, producing large amounts of carbonaceous sludge, acid, water, and other contaminants. In addition,
a burnout can chemically alter the compressor lubricant, and/or
thermally decompose refrigerant in the vicinity of the burn. Products of burnout escape into the system, causing severe cleanup problems. If decomposition products are not removed, replacement
motors fail with increasing frequency.
Although the Refrigeration Service Engineers Society (RSES
1988) differentiates between mild and severe burnouts, many compressor manufacturers service bulletins treat all burnouts alike. A
rapid burn from a spot failure in the motor winding results in a mild
burnout with little lubricant discoloration and no carbon deposits. A
severe burnout occurs when the compressor remains online and
burns over a longer period, resulting in highly discolored lubricant,
carbon deposits, and acid formation.
Because the condition of the lubricant can be used to indicate
the amount of contamination, the lubricant should be examined
during the cleanup process. Wojtkowski (1964) stated that acid in
R-22/mineral oil systems should not exceed 0.05 total acid number
(mg KOH per g oil). Commercial acid test kits can be used for this
analysis. An acceptable acid number for other lubricants has not
been established.
Various methods are recommended for cleaning a system after
hermetic motor burnout (RSES 1988). However, the suction-line
filter-drier method is commonly used (see the section on System
Cleanup Procedure after Hermetic Motor Burnout).

Field Assembly
Proper field assembly and maintenance are essential for contaminant control in refrigerating systems and to prevent undesirable refrigerant emissions to the atmosphere. Driers may be too
small or carelessly handled so that drying capacity is lost. Improper tube-joint soldering is a major source of water, flux, and
oxide scale contamination. Copper oxide scale from improper
brazing is one of the most frequently observed contaminants.
Careless tube cutting and handling can introduce excessive quantities of dirt and metal chips. Take care to minimize these sources
of internal contamination. In addition, because an assembled system cannot be dehydrated easily, oversized driers should be installed. Even if components are delivered sealed and dry, weather
and the amount of time the unit is open during assembly can introduce large amounts of moisture.
In addition to internal sources, external factors can cause a unit to
fail. Too small or too large transport tubing, mismatched or misapplied components, fouled air condensers, scaled heat exchangers,
inaccurate control settings, failed controls, and improper evacuation
are some of these factors.

SYSTEM CLEANUP PROCEDURE AFTER

HERMETIC MOTOR BURNOUT
This procedure is limited to positive-displacement hermetic compressors. Centrifugal compressor systems are highly specialized and

are frequently designed for a particular application. A centrifugal

system should be cleaned according to the manufacturers recommendations. All or part of the procedure can be used, depending
on factors such as severity of burnout and size of the refrigeration
system.
After a hermetic motor burnout, the system must be cleaned thoroughly to remove all contaminants. Otherwise, a repeat burnout will
likely occur. Failure to follow these minimum cleanup recommendations as quickly as possible increases the potential for repeat
burnout.

Procedure
A. Make sure a burnout has occurred. Although a motor that will
not start appears to be a motor failure, the problem may be
improper voltage, starter malfunction, or a compressor mechanical fault (RSES 1988). Investigation should include the following steps:
1. Check for proper voltage.
2. Check that the compressor is cool to the touch. An open internal overload could prevent the compressor from starting.
3. Check the compressor motor for improper grounding using a
megohmmeter or a precision ohmmeter.
4. Check the external leads and starter components.
5. Obtain a small sample of oil from the compressor, examine it
for discoloration, and analyze it for acidity.
B. Safety. In addition to electrical hazards, service personnel
should be aware of the hazard of acid burns. If the lubricant or
sludge in a burned-out compressor must be touched, wear rubber
gloves to avoid a possible acid burn.
C. Cleanup after a burnout. Just as proper installation and service
procedures are essential to prevent compressor and system failures, proper system cleanup and installation procedures when
installing the replacement compressor are also essential to prevent repeat failures. Key elements of the recommended procedures are as follows:
1. U.S. federal regulations require that the refrigerant be isolated in the system or recovered into an external storage container to avoid discharge into the atmosphere. Before
opening any portion of the system for inspection or repairs,
refrigerant should be recovered from that portion until the
vapor pressure reduces to less than 12.7 psia (4 in. Hg vacuum) for R-22 or 9.8 psia (10 in. Hg vacuum) for CFC or
other HCFC systems.
2. Remove the burned-out compressor and install the replacement. Save a sample of the new compressor lubricant that has
not been exposed to refrigerant and store in a sealed glass bottle. This will be used later for comparison.
3. Inspect all system controls such as expansion valves, solenoid valves, check valves, etc. Clean or replace if necessary.
4. Install an oversized drier in the suction line to protect the
replacement compressor from any contaminants remaining
in the system. Install a pressure tap upstream of the filterdrier, to allow measuring the pressure drop from tap to service valve during the first hours of operation to determine
whether the suction line drier needs to be replaced.
5. Remove the old liquid-line drier, if one exists, and install a
replacement drier of the next larger capacity than is normal
for this system. Install a moisture indicator in the liquid line
if the system does not have one.
6. Evacuate and leak-check the system or portion opened to the atmosphere according to the manufacturers recommendations.
7. Recharge the system and begin operations according to the
manufacturers start-up instructions, typically as follows:
a. Observe pressure drop across the suction line drier for the
first 4 h. Follow the manufacturers guide; otherwise,

Control of Moisture and Other Contaminants in Refrigerant Systems

compare to pressure drop curve in Figure 7 and replace
driers as required.
b. After 24 to 48 h, check pressure drop and replace driers as
required. Take a lubricant sample and check with an acid
test kit. Compare the lubricant sample to the initial sample
saved at the time the replacement compressor was
installed. Cautiously smell the lubricant sample. Replace
lubricant if acidity persists or if color or odor indicates.
c. After 7 to 10 days or as required, repeat step b.
D. Additional suggestions
1. If sludge or carbon has backed up into the suction line, swab
it out or replace that section of the line.
2. If a change in the suction line drier is required, change the
lubricant in the compressor each time the cores are changed,
if compressor design permits.
3. Remove the suction line drier after several weeks of system
operation to avoid excessive pressure drop in the suction line.
This problem is particularly significant on commercial refrigeration systems.
4. Noncondensable gases may be produced during burnout. With
the system off, compare the head pressure to the saturation
pressure after stabilization at ambient temperature. Adequate
time must be allowed to ensure stabilization. If required,
purge the charge by recycling it or submit the purged material
for reclamation.

Special System Characteristics and Procedures

Because of unique system characteristics, the procedures described here may require adaptations.
A. If a lubricant sample cannot be obtained from the new compressor, find another way to get a sample from the system.
1. Install a tee and a trap in the suction line. An access valve at
the bottom of the trap permits easy lubricant drainage. Only
0.5 oz of lubricant is required for an acid analysis. Be certain the lubricant sample represents lubricant circulating in
the system. It may be necessary to drain the trap and discard the first amount of lubricant collected, before collecting the sample to be analyzed.
LIVE GRAPH

Click here to view

Fig. 7

Fig. 7

Maximum Recommended Filter-Drier Pressure Drop

Maximum Recommended Filter-Drier Pressure Drop

6.9

2. Make a trap from 1 3/8 in. copper tubing and valves. Attach
this trap to the suction and discharge gage port connections
with a charging hose. By blowing discharge gas through the
trap and into the suction valve, enough lubricant will be collected in the trap for analysis. This trap becomes a tool that
can be used repeatedly on any system that has suction and
discharge service valves. Be sure to clean the trap after every
use to avoid cross contamination.
B. On semihermetic compressors, remove the cylinder head to
determine the severity of burnout. Dismantle the compressor for
solvent cleaning and hand wiping to remove contaminants.
Consult the manufacturers recommendations on compressor
rebuilding and motor replacement.
C. In rare instances on a close-coupled system, where it is not feasible to install a suction line drier, the system can be cleaned by
repeated changes of the cores in the liquid line drier and repeated
lubricant changes.
D. On heat pumps, the four-way valve and compressor should be
carefully inspected after a burnout. In cleaning a heat pump after
a motor burnout, it is essential to remove any drier originally
installed in the liquid line. These driers may be replaced for
cleanup, or a biflow drier may be installed in the common reversing liquid line.
E. Systems with a critical charge require a particular effort for
proper operation after cleanup. If an oversized liquid-line drier
is installed, an additional charge must be added. Check with
the drier manufacturer for specifications. However, no additional charge is required for the suction line drier that may be
added.
F. The new compressor should not be used to pull a vacuum. Refer
to the manufacturers recommendations for evacuation. Normally, the following method is used, after determining that there
are no refrigerant leaks in the system:
a. Pull a high vacuum to an absolute pressure of less than
500 m Hg for several hours.
b. Allow the system to stand several hours to be sure the vacuum is maintained. This requires a good vacuum pump and
an accurate high-vacuum gage.

CONTAMINANT CONTROL DURING RETROFIT

Because of the phaseout of CFCs, existing refrigeration and airconditioning systems are commonly retrofitted to alternative refrigerants. The term refrigerant in this section refers to a fluorocarbon
working fluid offered as a possible replacement for a CFC, whether
that replacement consists of one chemical, an azeotrope of two
chemicals, or a blend of two or more chemicals. The terms retrofitting and conversion are used interchangeably to mean the
modification of an existing refrigeration or air-conditioning system
designed to operate on a CFC so that it can safely and effectively
operate on an HCFC or HFC refrigerant. This section only covers
the contaminant control aspects of such conversions. Equipment
manufacturers should be consulted for guidance regarding the specifics of actual conversion. Industry standards and manufacturers
literature are also available that contain supporting information
(e.g., UL Standards 2170, 2171, and 2172).
Contaminant control concerns for retrofitting a CFC system to an
alternative refrigerant fall into the following categories:
Cross-contamination of old and new refrigerants. This should
be avoided even though there are usually no chemical compatibility problems between the CFCs and their replacement refrigerants. One problem with mixing refrigerants is that it is difficult to
determine system performance after retrofit. Pressure/temperature
relationships are different for a blend of two refrigerants than for
each refrigerant individually. A second concern with mixing

6.10

2006 ASHRAE HandbookRefrigeration

refrigerants is that if the new refrigerant charge must be removed

in the future, the mixture may not be reclaimable (DuPont 1992).
Cross-contamination of old and new lubricant. Equipment
manufacturers generally specify that the existing lubricant be
replaced with the lubricant they consider suitable for use with a
given HFC refrigerant. In some cases, the new lubricant is incompatible with the old one or with chlorinated residues present. In
other cases, the old lubricant is insoluble with the new refrigerant
and tends to collect in the evaporator, interfering with heat transfer. For example, when mineral oil is replaced by a polyol ester
lubricant during retrofit to an HFC refrigerant, a typical recommendation is to reduce the old oil content to 5% or less of the
nominal oil charge (Castrol 1992). Some retrofit recommendations specify lower levels of acceptable contamination for polyol
ester lubricant/HFC retrofits, so original equipment manufacturers recommendations should be obtained before attempting a
conversion.
Chemical compatibility of old system components with new
fluids. One of the preparatory steps in a retrofit is to confirm that
either the existing materials in the system are acceptable or that
replacement materials are on hand to be installed in the system
during the retrofit. Fluorocarbon refrigerants generally have solvent properties, and some are very aggressive. This characteristic can lead to swelling and extrusion of polymer O rings,
undermining their sealing capabilities. Material can also be
extracted from polymers, varnishes, and resins used in hermetic
motor windings. These extracts can then collect in expansion
devices, interfering with system operation. Residual manufacturing fluids such as those used to draw wire for compressor motors
can be extracted from components and deposited in areas where
they can interfere with operation. Suitable materials of construction have been identified by equipment manufacturers for use
with HFC refrigerant systems.
Drier media must also be chemically compatible with the new
refrigerant and effective in removing moisture, acid, and particulate in the presence of the new refrigerant. Drier media commonly
used with CFC refrigerants tend to accept small HFC refrigerant
molecules and lose moisture retention capability (Cohen and
Blackwell 1995). Drier media have been developed that minimize
this tendency.

CHILLER DECONTAMINATION
Chiller decontamination is used to clean reciprocating, rotary
screw, and centrifugal machines. Large volumes of refrigerant are
circulated through a contaminated chiller while continuously
being reclaimed. It has been used successfully to restore many
chillers to operating specifications. Some chillers have been saved
from early retirement by decontamination procedures. Variations
of the procedure are myriad and have been used for burnouts,
water-flooded barrels, particulate incursions, chemical contamination events, brine leaks, and oil strips. One frequently used technique is to perform numerous batch cycles, thus increasing the
velocity-based cleansing component. Excess oil is stripped out to
improve chiller heat transfer efficiency. The full oil charge can be
removed in preparation for refrigerant conversion.
Low-pressure units require different machinery than high-pressure
units. It is best to integrate decontamination and mechanical
services early into one overall procedure. On machines that require
compressor rebuild, it is best to perform decontamination work while
the compressor is removed or before it is rebuilt, particularly for
reciprocating units. Larger-diameter or relocated access ports may
be requested. The oil sump will be drained. For chillers that cannot
be shut down, special online techniques have been developed using
reclamation. The overall plan is coordinated with operations personnel to prevent service interruptions. For some decontamination

projects, it is advantageous to have the water boxes open; in other

cases, closed. Intercoolers offer special challenges.

REFERENCES
ARI. 1999. Appendix C to ARI Standard 700Analytical procedures for
ARI Standard 700. Standard 700c-1999. Air-Conditioning and Refrigeration Institute, Arlington, VA.
ARI. 2004. Liquid-line driers. Standard 710-04. Air-Conditioning and
Refrigeration Institute, Arlington, VA.
ARI. 2001. Flow-capacity rating and application of suction-line filters and
filter-driers. Standard 730-01. Air-Conditioning and Refrigeration Institute, Arlington, VA.
ASHRAE. 1992. Method of testing desiccants for refrigerant drying. Standard 35-1992.
ASHRAE. 1995. Method of testing liquid line refrigerant driers. ANSI/
ASHRAE Standard 63.1-1995 (RA 01).
ASHRAE. 1996. Method of testing liquid line filter-drier filtration capability. ANSI/ASHRAE Standard 63.2-1996.
ASHRAE. 1985. Method of testing flow capacity of suction line filter driers.
ANSI/ASHRAE Standard 78-1985 (RA 03).
ASHRAE. 2002. Reducing the release of halogenated refrigerants from
refrigerating and air-conditioning equipment and systems. ANSI/
ASHRAE Standard 147-2002.
ASTM. 2003. Standard specification for seamless copper tube for air conditioning and refrigeration field service. Standard B280-03. American
Society for Testing and Materials, West Conshohocken, PA.
Brisken, W.R. 1955. Moisture migration in hermetic refrigeration systems as
measured under various operating conditions. Refrigerating Engineering
(July):42.
Castrol. 1992. Technical Bulletin 2. Castrol Industrial North America,
Specialty Products Division, Irvine, CA.
Cavestri, R.C. and W.R. Schafer. 1999. Equilibrium water capacity of desiccants in mixtures of HFC refrigerants and appropriate lubricants.
ASHRAE Transactions 104(2):60-65.
Cavestri, R.C. and D.L. Schooley. 1998. Test methods for inorganic acid
removal capacity in desiccants used in liquid line driers. ASHRAETransactions 104(1B):1335-1340.
Cohen, A.P. 1993. Test methods for the compatibility of desiccants with
alternative refrigerants. ASHRAE Transactions 99(1):408-412.
Cohen, A.P. 1994. Compatibility and performance of molecular sieve desiccants with alternative refrigerants. Proceedings of the International Conference: CFCs, The Day After. International Institute of Refrigeration,
Paris.
Cohen, A.P. and C.S. Blackwell. 1995. Inorganic fluoride uptake as a measure of relative compatibility of molecular sieve desiccants with fluorocarbon refrigerants. ASHRAE Transactions 101(2):341-347.
Cohen, A.P. and S.R. Dunne. 1987. Review of automotive air-conditioning
drydown rate studiesThe kinetics of drying Refrigerant 12. ASHRAE
Transactions 93(2):725-735.
DeGeiso, R.C. and R.F. Stalzer. 1969. Comparison of methods of moisture
determination in refrigerants. ASHRAE Journal (April).
Doderer, G.C. and H.O. Spauschus. 1966. A sealed tube-gas chromatograph
method for measuring reaction of Refrigerant 12 with oil. ASHRAE
Transactions 72(2):IV.4.1-IV.4.4.
Dunne, S.R. and T.J. Clancy. 1984. Methods of testing desiccant for refrigeration drying. ASHRAE Transactions 90(1A):164-178.
DuPont. 1992. Acceptance specification for used refrigerants. Bulletin H31790-1. E.I. DuPont de Nemours and Company, Wilmington, DE.
Elsey, H.M. and L.C. Flowers. 1949. Equilibria in Freon-12Water systems. Refrigerating Engineering (February):153.
Elsey, H.M., L.C. Flowers, and J.B. Kelley. 1952. A method of evaluating
refrigerator oils. Refrigerating Engineering (July):737.
Griffith, R. 1993. Polyolesters are expensive, but probably a universal fit. Air
Conditioning, Heating & Refrigeration News (May 3):38.
Gully, A.J., H.A. Tooke, and L.H. Bartlett. 1954. Desiccant-refrigerant
moisture equilibria. Refrigerating Engineering (April):62.
Gustafsson, V. 1977. Determining the air content in small refrigeration systems. Purdue Compressor Technology Conference.
Haagen-Smit, I.W., P. King, T. Johns, and E.A. Berry. 1970. Chemical
design and performance of an improved Karl Fischer titrator. American
Laboratory (December).
Hoffman, J.E. and B.L. Lange. 1962. Acid removal by various desiccants.
ASHRAE Journal (February):61.

Control of Moisture and Other Contaminants in Refrigerant Systems

Jones, E. 1969. Liquid or suction line drying? Air Conditioning and Refrigeration Business (September).
Krause, W.O., A.B. Guise, and E.A. Beacham. 1960. Time factors in the
removal of moisture from refrigerating systems with desiccant type driers. ASHRAE Transactions 66:465-476.
Kvalnes, D.E. 1965. The sealed tube test for refrigeration oils. ASHRAE
Transactions 71(1):138-142.
Mays, R.L. 1962. Molecular sieve and gel-type desiccants for Refrigerants
12 and 22. ASHRAE Journal (August):73.
Norton, F.J. 1957. Rates of thermal decomposition of CHClF2 and CF2Cl2.
Refrigerating Engineering (September):33.
Parmelee, H.M. 1965. Sealed tube stability tests on refrigeration materials.
ASHRAE Transactions 71(1):154-161.
Rosen, S., A.A. Sakhnovsky, R.B. Tilney, and W.O. Walker. 1965. A method
of evaluating filtration and flow characteristics of liquid line driers.
ASHRAE Transactions 71(1):200-205.
RSES. 1988. Standard procedure for replacement of components in a sealed
refrigerant system (compressor motor burnout). Refrigeration Service
Engineers Society, Des Plaines, IL.
Spauschus, H.O. and G.C. Doderer. 1961. Reaction of Refrigerant 12 with
petroleum oils. ASHRAE Journal (February):65.
Spauschus, H.O. and G.C. Doderer. 1964. Chemical reactions of Refrigerant
22. ASHRAE Journal (October):54.
Spauschus, H.O. and R.S. Olsen. 1959. Gas analysisA new tool for determining the chemical stability of hermetic systems. Refrigerating Engineering (February):25.
Thrasher, J.S., R. Timkovich, H.P.S. Kumar, and S.L. Hathcock. 1993. Moisture solubility in Refrigerant 123 and Refrigerant 134a. ASHRAE Transactions 100(1):346-357.
UL. 2001. Standard for refrigerant-containing components and accessories,
nonelectrical. ANSI/UL Standard 207-01. Underwriters Laboratories,
Northbrook, IL.
UL. 1993. Field conversion/retrofit of products to change to an alternative
refrigerantConstruction and operation. Standard 2170-93. Underwriters Laboratories, Northbrook, IL.
UL. 1993. Field conversion/retrofit of products to change to an alternative
refrigerantInsulating material and refrigerant compatibility. Standard
2171-93. Underwriters Laboratories, Northbrook, IL.
UL. 1993. Field conversion/retrofit of products to change to an alternative
refrigerantProcedures and methods. Standard 2172-93. Underwriters
Laboratories, Northbrook, IL.
Walker, W.O. 1963. Latest ideas in use of desiccants and driers. Refrigerating Service & Contracting (August):24.
Walker, W.O., J.M. Malcolm, and H.C. Lynn. 1955. Hydrophobic behavior
of certain desiccants. Refrigerating Engineering (April):50.

6.11

Walker, W.O., S. Rosen, and S.L. Levy. 1962. Stability of mixtures of refrigerants and refrigerating oils. ASHRAE Journal (August):59.
Wojtkowski, E.F. 1964. System contamination and cleanup. ASHRAE Journal (June):49.

BIBLIOGRAPHY
Boing, J. 1973. Desiccants and driers. RSES Service Manual, Section 5, 62016B. Refrigeration Service Engineers Society, Des Plaines, IL.
Burgel, J., N. Knaup, and H. Lotz. 1988. Reduction of CFC-12 emission
from refrigerators in the FRG. International J ournal of Refrigeration
11(4).
Byrne, J.J., M. Shows, and M.W. Abel. 1996. Investigation of flushing and
clean-out methods for refrigeration equipment to ensure system compatibility. Air-Conditioning and Refrigeration Technology Institute, Arlington, VA. DOE/CE/23810-73.
DuPont. 1976. Mutual solubilities of water with fluorocarbons and fluor ocarbon-hydrate formation. DuPont de Nemours and Company, Wilmington, DE.
Guy, P.D., G. Tompsett, and T.W. Dekleva. 1992. Compatibilities of nonmetallic materials with R-134a and alternative lubricants in refrigeration
systems. ASHRAE Transactions 98(1):804-816.
Jones, E. 1964. Determining pressure drop and refrigerant flow capacities of
liquid line driers. ASHRAE Journal (February):70.
Kauffman, R.E. 1992. Sealed tube tests of refrigerants from field systems
before and after recycling. ASHRAE Transactions 99(2):414-424.
Kitamura, K., T. Ohara, S. Honda, and H. SakaKibara. 1993. A new
refrigerant-drying method in the automotive air conditioning system
using HFC-134a. ASHRAE Transactions 99(1):361-367.
Manz, K.W. 1988. Recovery of CFC refrigerants during service and recycling by the filtration method. ASHRAE Transactions 94(2):2145-2151.
McCain, C.A. 1991. Refrigerant reclamation protects HVAC equipment
investment. ASHRAE Journal 33(4).
Sundaresan, S.G. 1989. Standards for acceptable levels of contaminants in
refrigerants. In CFCsTime of Transition, pp. 220-223. ASHRAE.
Walker, W.O. 1960. Contaminating gases in refrigerating systems. In RSES
Service Manual, Section 5, 620-15. Refrigeration Service Engineers
Society, Des Plaines, IL.
Walker, W.O. 1985. Methyl alcohol in refrigeration. In RSES Service Manual, Section 5, 620-17A. Refrigeration Service Engineers Society, Des
Plaines, IL.
Zahorsky, L.A. 1967. Field and laboratory studies of wax-like contaminants in
commercial refrigeration equipment. ASHRAE Transactions 73(1):II.1.1II.1.9.
Zhukoborshy, S.L. 1984. Application of natural zeolites in refrigeration
industry. Proceedings of the International Symposium on Zeolites,
Portoroz, Yugoslavia (September).