THE EFFECTS OF

STATIC HEAD
in Refrigeration Systems
Not considered carefully during design,
pressure exerted by liquid can have
a profound impact on system operation

n refrigerating systems, static head is extremely out of the end of the pipe. It would rise to an

I important, bearing upon the performance elevation of 30 ft, where it would remain. A gauge
of the system in many ways. Sometimes, placed on the pipe at the water level would read
designers fail to appreciate the effects of static head 0 psig, or atmospheric pressure, while a gauge
when sizing system components placed at the pump discharge would
and selecting operating conditions. By RONALD A. COLE, PE read approximately 13 psig, the
This article will focus on static R.A. Cole & Associates pressure required to lift the water
head as it applies to large, custom- Seattle, Wash. 30 ft.
designed and built ammonia
refrigerating systems, as well as the principles that STATIC HEAD IN REFRIGERANT LIQUID
must be considered in design. Unlike water, such as that in chilled-water or
glycol systems, the working fluid in refrigerating
STATIC HEAD DEFINED systems is volatile—that is, it operates at or near
Static head is the pressure exerted by a pool of its boiling point. With few exceptions, unless steps
liquid. Because of the weight of the liquid, this are taken to cause the fluid to be subcooled, most
pressure is greater at the bottom of the pool than commonly, it will be saturated. That means the
at the top; so, the deeper the pool, the greater temperature of the liquid will be its boiling point.
the pressure at the bottom. Despite its name,
static head exists whether liquid is stationary or in
Gauge: 0 psig
motion.
Static head, along with dynamic head, is the Water level
force against which pumps must work to transport
liquid from the physical bottom of a system to the
physical top. Thus, the capacity of a pump is
expressed as a liquid flow of a certain number of
gallons per minute at a head of a certain number of 32 ft
feet.
30 ft
If the static head of a system exceeds the rated
head of the system’s pump, liquid will not be lifted
to the top of the system. For example, in Figure 1,
the pump, with a maximum discharge head rating
of 30 ft, is intended to move water out of the Gauge: 13 psig
reservoir, transport it up the vertical pipe, and spray
it out of the end of the pipe. If the height of the pipe
were greater than 30 ft, the water would not spray
Reservoir

Pump
A member of HPAC Engineering’s Editorial Advi-
sory Board, Ronald A. Cole, PE, is president of R.A. FIGURE 1. The effect of static head on a water-
Cole & Associates. pumping system.

HPAC Engineering • September 2001 53

 S T A T I C H E A D

the liquid is distributed to the plant. On purposes of this discussion, flow friction,
Liquid to plant the other hand, with a receiver protected heat transfer, and the effects of changes
from heat sources and with a cool in “superficial” density, approximately
ambient, some amount of subcooling 0.6 percent by weight of the flowing liq-
could be obtained in the receiver. uid refrigerant will turn to gas. This
Prudence dictates, however, that unless number is very small and appears to be
30 ft Condensed-liquid drain
measures are taken during design to inconsequential. However, with a low-
ensure that there is subcooling of the molecular-weight fluid such as ammo-
Equalizing liquid, it should be assumed that the nia, the gas is very light. That amount of
connection refrigerant liquid in the receiver is flash gas would be 38 percent by volume
saturated. of the flowing fluid. In other words, the
Figure 2 depicts a typical high- volume would increase by 38 percent.
pressure liquid receiver found in a One adverse effect of the formation of
Liquid large refrigerating system. Absent flash gas is the increased velocity at which
level
extraordinary conditions, the pressure the liquid-gas mixture flows because of
in the receiver would be the high-side the increase in volume. That increased
FIGURE 2. Typical high-pressure liquid
pressure of the system—that is, it velocity significantly increases the
receiver and piping in large refrigerating
would be very close to the condensing frictional pressure loss associated with the
system.
pressure—especially if the receiver flow. However, if, in the example in Fig-
was separately equalized to the high-side ure 2, a liquid subcooler is installed in the
piping.1 With the liquid in the receiver liquid line (Figure 3) and sized to cool the
assumed to be saturated, the system liquid exiting the receiver by 3 F, ignoring
Liquid to plant
pressure in the receiver would be the other heat transfer, the temperature of
Liquid subcooler vapor pressure of the refrigerant. the liquid at the top of the vertical run of
The pressure in the receiver in Figure 2 the pipe would be 88.4 F, and the vapor
is determined by the condensing pressure of the liquid would be 176.2
temperature. The pressure at the psia, lower than the actual liquid pres-
Condensed-liquid drain
interface, where the liquid and vapor are sure. As a result, no flash gas would form.
30 ft
in equilibrium, is the greatest pressure Case study. At a grocery-distribution
Equalizing that can exist in the receiver. Because the warehouse in Illinois, all of the cooling
connection inlet to the liquid line is at the bottom units in the storage rooms are arranged
of a pool of liquid, it is subject to the for liquid recirculation, while five units
same pressure variation from static head on a refrigerated dock are arranged for
described earlier. However, because that direct expansion—that is, liquid feed
Liquid pressure is the highest pressure that can supplied by a thermostatic expansion
level
exist in the receiver, there is a pressure
gradient in the riser, decreasing along the
FIGURE 3. Liquid receiver with a liquid length of the vertical run of the pipe so
100 ft through 150 ft through
subcooler. that the pressure at the top of the pipe is ambient space 36-F space
lower than that in the receiver by the
One exception to this rule occurs in amount of the static head.
certain air-cooled condenser systems, If the high-side pressure in the system,
where circuiting may be arranged to as well as the vapor pressure of the refrig-
allow subcooling of the high-pressure erant in the receiver, in Figure 2 were 185
liquid. But in large industrial systems, psia and the refrigerant were ammonia,
where condensers typically are of the the temperature of the liquid in the re-
evaporative type, there is little or no ceiver would be 91.4 F. With a 30-ft ver-
subcooling in the liquid exiting the tical rise in the pipe coming from the re-
condenser. Most large systems are ceiver, the theoretical pressure of the
installed with high-pressure liquid liquid in the pipe would be 177.3 psia at Dock-cooling
units
receivers to allow storage of refrigerant the top, based on a liquid density of
and to accommodate the effects of 36.87 lb per cu ft. Liquid receiver
system transients, or surges. In these The saturation temperature of ammo-
cases, even if subcooling were imparted nia corresponding to a vapor pressure of
to the liquid in the condenser, depending 177.3 psia is 88.8 F. In effecting 2.6-F
on system conditions, ambient condi- cooling, as the pressure of the liquid FIGURE 4. Schematic of high-pressure
tions, and the location of the receiver, decreases along the vertical path of the liquid piping at grocery-distribution
that subcooling could be lost before pipe, flash gas forms. Ignoring, for the warehouse in Illinois.

54 September 2001 • HPAC Engineering

 S T A T I C H E A D

valve. This was done to avoid the effects liquid at the lower temperature and the Ensuring bubble-free liquid at the
of static head on the suction line (to be variable relationship between vapor cooling units required subcooling suffi-
discussed later in this article), which also pressure and temperature. Moreover, the cient enough to offset the effects of static
can affect the operation of the cooling amount of cooling that can be obtained head, friction, and ambient heat transfer.
units. To ensure adequate performance of by passing the uninsulated pipe through This was accomplished using a cold pipe
the thermostatic expansion valves, it is the storage cooler is decreased because running in the same pipe rack as the
imperative that there be no flash gas in of the smaller temperature difference liquid line in question. The cold pipe was
the liquid at the inlet to the valve. between the refrigerant liquid and the air. a two-phase return line that was part of
Figure 4 shows the piping and
plant layout of the warehouse. The
liquid line rises vertically for 30 ft, runs
horizontally through ambient space for
approximately 100 ft, runs through a
refrigerated room for 150 ft, and then
drops vertically approximately 22 ft to
feed the five dock units, which are
located under a mezzanine. As initially
proposed, there would have been no
subcooling of this liquid. The engineer
in charge of design contended that the
liquid would be sufficiently subcooled by
passage of the uninsulated pipe through
the 36-F cooler and that the static head
would be of little, if any, import because
the up-run and the down-run would, at
least in part, cancel each other out.
There were two flaws in the engineer’s
reasoning. First, the amount of subcool-
ing obtained by passing the liquid line
through the refrigerated space may not
have been sufficient under all conditions,
if it was sufficient at all. Second, the static
heads would not necessarily have
canceled each other out. Once flash gas
has formed in a vertical rise, merely
turning the pipe downward does not
eliminate the flash gas. The only way to
eliminate flash gas is to prevent it from
forming or condense it back to a liquid—
in either case—through a heat-transfer
(cooling) process.
A theoretical calculation of potential
heat transfer by natural convection from
liquid in a 1-in. pipe to the air in the
storage cooler suggests that the liquid
possibly could be cooled by 2.8 F at
design conditions (i.e., 95-F condensing
temperature), an amount sufficient to
offset the effects of the vertical lift
illustrated in Figure 4. However, this
does not take into account friction in the
pipe, which also reduces the pressure
along the flow path, and any effects of
heat transfer that might warm the liquid
as it flows through ambient space.
At a liquid temperature of 85 F,
however, the effect of static head alone is
2.9 F, owing to the greater density of the

HPAC Engineering • September 2001 55

 S T A T I C H E A D

the liquid-recirculation system for the cooling units in the stor-

age coolers. Because there was cold liquid in this pipe already, a
dam was created in a short section of the pipe, and the liquid
line was directed so that it flowed through the pool of cold
liquid. The short section of return line with the subcooling pipe
in it is shown in Figure 5. The arrangement and interconnec-
tion of the return
line and liquid
line are shown in
Figure 6. Approx-
imately 8 F of liq-
uid subcooling—
enough to ensure
bubble-free liquid
Eccentric 1-in. pipe inside at the thermosta-
reducers tic expansion
valves under all
FIGURE 5. Field-fabricated liquid subcooler conditions—was
used in grocery-distribution warehouse in obtained using
Illinois. this method.

STATIC HEAD AND EVAPORATORS

The most troublesome consequences of static head concern
the operation of evaporators, particularly at low temperatures.
In large refrigerating plants and many process applications, the
most common method of supplying liquid to evaporators
is “liquid overfeed,” by which refrigerant is pumped to an
evaporator at a low pressure and at a rate of flow that exceeds
the evaporation rate. The reason for this is to ensure full wetting
of the heat-transfer surfaces of the evaporator. Thus, the
suction, or return, line carries a mixture of liquid and vapor. A
number of important factors must be considered when
designing piping for this type of application, the most
important being the excess pressure drop associated with the
liquid that is carried along the pipe with the gas.
Another critical factor affecting the operation of these
kinds of piping systems concerns the elevated positioning of
piping runs in a plant or—even more commonly today—on
the roof of a plant. With many types of refrigerated processing
equipment—such as plate freezers, continuous belt-type
coolers and freezers, and some chillers—located at floor level,
a vertical section of return pipe is necessary at some point in
the run. The problem is getting the liquid to flow vertically in
the riser without severely affecting the operating conditions of
the evaporator.
Figure 7 illustrates this problem. A plate freezer is intended
to operate at a temperature of -45 F. It has a suction pipe that
must run vertically for 20 ft to get to the overhead mains that
run back to the machine room. The liquid-feed arrangement
is overfeed, so there is both liquid and gas in the suction line,
including in the riser. Unless the riser is properly designed to
ensure a flow condition that enables the gas to carry the liquid
up the riser, the liquid will fall to the bottom of the pipe and
build up in a pool until it fills the pipe entirely. The gas then will
alternately bubble up through the liquid in the vertical section
of the pipe and “burp” slugs of liquid up the pipe, much like the
action of the water in a percolator-type coffee pot.
The effect of such a liquid hold-up condition on the

HPAC Engineering • September 2001 57

 S T A T I C H E A D

operation of a freezer is dramatic. The worst-case scenario is the

vertical section of pipe filling full of liquid. With ammonia,
at -45 F, the static pressure caused by 20 ft of liquid is 6 psi. The
vapor pressure at -45 F is 8.9 psia. Thus, at the outlet from the
freezer, the pressure is 14.9 psia. The saturation temperature at
the freezer outlet is -27.5 F. A 17.5-F loss in operating-tempera-
ture difference is
suffered because of
Liquid to dock units the effects of static
head in the return
line. In other words,
to maintain an
evaporator temper-
ature of -27.5 F, it is
necessary to operate
Subcooler the compressor at a
suction temperature
less than -45 F to
compensate for the
20-F two-phase return from plant
static-head loss.
If the refrigerant
Liquid from receiver were R-22, the dif-
ference would be
even more dra-
FIGURE 6. Installation detail of subcooler in matic. The pressure
grocery-distribution warehouse in Illinois. caused by static
head would be 12.3
psi, and the temperature penalty would be more than 26 F.
In other words, the temperature at the freezer outlet would be
-19 F, with a temperature in the overhead main of -45 F.
These are not hypothetical situations. Measurements of
losses in risers caused by the effects of static head have been
made in food-processing plants in Ohio, Illinois, Tennessee,
and North Carolina. It was found that the pressure loss in the
risers in these plants was nearly equal to the static head of liquid
refrigerant at that temperature. Richards2 suggested a method
of sizing pipes that minimizes the possibility of stacking up
liquid in risers and, thus, avoids the extreme static-head penalty.
However, in cases in which piping was not properly designed
or there were varying
conditions, a common
-45 F return main
solution to the problem
was to use some kind of
pumping system to
Return
pump the liquid up to the
line 20 ft
higher elevation while
allowing the gas to flow in
a “dry” suction line—that
is, without any liquid—
in the normal manner.
Finally, a very com-
Plate
freezer
mon method of feeding
evaporators is “flooding”
them. One configuration
of flooded evaporator in-
FIGURE 7. The effect of static head on corporates a surge drum.
the return pipe of a low-temperature Examples of evaporators
process freezer. that commonly utilize

HPAC Engineering • September 2001 59

 S T A T I C H E A D

the surge drum are air-cooling coils, pipe-coil ice builders, plate
and frame chillers, and numerous types of process coolers
and freezers. The location of the surge drum relative to the
evaporator is crucial to the success of the application.
Figure 8 is a schematic of a flooded evaporator with a surge
drum. In many cases, physical constraints, space limitations, or
other factors result in the surge drum being mounted a great
distance above the evaporator. The static-head-penalty problem
discussed earlier once again comes into play. If the evaporator
in this case is intended to operate at a temperature of -40 F,
and the surge drum is located 12 ft above the evaporator, the
pressure at the inlet to the evaporator will be 3.6-psi higher than
the pressure in the surge drum, if the refrigerant is ammonia.
That means that the evaporating temperature at the inlet to the
evaporator will be 10.2-F higher than the actual liquid temper-
ature. Again, were the refrigerant R-22, the differences would
be even greater.
For it to begin to boil,
Surge drum the liquid must be
warmed that 10.2 F. In
other words, even
though the surge drum is
operating at a tempera-
ture of -40 F, and the
Static
compressor is operating
head at a suction temperature
commensurate with
that, the evaporator will
be operating at a temper-
Flooded ature at least 10-F higher.
cooling coil Of course, everything
else in the cooling
FIGURE 8. Flooded cooling unit with process associated with
surge drum mounted very high above it. the evaporator also will
be operating at a corre-
spondingly higher temperature. All of this can be attributed to
the static head of the refrigerant liquid. That is why it is
recommended that surge drums be located very close to the
evaporator and that the head allowed for in the vertical height
of the surge drum be only that required for the flow friction of
getting the liquid into and through the evaporator and getting
the gas-liquid mixture returned to the surge drum.

CONCLUSION
As long as elements of refrigerating systems are located at
different elevations, the effects of static head will impose
themselves on system operation. It is imperative that designers
and operators of refrigerating plants understand what these
effects are and how they impact the operation of systems and
take whatever measures are necessary to ensure that systems
perform as intended.

REFERENCES
1) Cole, R.A. (1986, July). Avoiding refrigeration condenser
problems: I. Heating/Piping/Air Conditioning, pp. 97-108.
2) Richards, W.V. (1982, March). Piping is piping ... or is it?
Technical Papers. International Institute of Ammonia Refrigera-
tion Fourth Annual Meeting.

HPAC Engineering • September 2001 61