1. INTRODUCTION

Steam is one of the most commonly used services in industry. It allows the energy in
the fuel burned in a boiler to be carried to some other point where it can provide
mechanical energy through an engine or, more commonly, simply heat.

Steam can be raised by burning fuel in a central boiler plant, by utilizing waste heat
from industrial processes, or it can be taken as a “product” from a back pressure or
extraction turbine used for power generation.

Whatever the source, an efficient steam distribution system is essential. To do its job
effectively, steam must be available at the required time in the right quantity, at the
right pressure and temperature.

This course and this manual are intended to provide operators of steam distribution
systems used for heating purposes with information and recommendations on how to
check these systems, to identify areas of energy waste, and to modify systems to reduce
losses. The course is intended to give practical advice, and, as such, this manual does
not discuss theoretical aspects of steam and steam usage. The approach has been to
provide as much information as possible in graphical format, assisting plant personnel
to make rapid calculations as a part of their daily work.

Chapters 2 and 3 provide a basic introduction to the parameters affecting energy

consumption in steam systems. Chapters 4 through 9 cover areas of operating
efficiency and energy conservation in steam systems, beginning with heat transfer in
steam systems, sizing of steam distribution systems, steam traps, condensate recovery,
and flash steam. Finally, Chapter 10 provides, a summary of steam system operating
efficiency parameters, and can be used as a checklist to evaluate the efficient operation
of any steam system.

 



2. SOME BASIC PRINCIPLES

2.1 INTRODUCTORY CONCEPTS

Steam is simply water in the vapor phase. It is simply produced by adding heat to
water. As heat is transferred to the water, at atmospheric pressure for example, the
temperature increases appreciably till it reaches abou t 1OO’C which is the boiling point
of the water at atmospheric pressure. At that temperature water is said to be saturated.
Any additional heat transfer results in a change of phase, that is some of the liquid
becomes vapor, and during this process both the temperature and pressure remain
constant. With the continual addition of heat all of the liquid vaporizes into steam,
which is said to be saturated steam. It is to be recalled that the amount of heat required
to convert a unit mass of saturated water to saturated steam is the latent heat of
vaporization of water. The saturation temperature designates the temperature at which
vaporization takes place at a given pressure, and this pressure is called the saturation
pressure for the given temperature , and there is a definite relation between the
saturation pressure and temperature . Exhibit 2.1 shows a typical curve for Such
relation.

If water exists at a temperature lower than the saturation temperature of the existing
pressure, it is said to be either Subcooled (implying that the temperature is less than the
saturation temperature for the given pressure) or compressed ( implying that the
pressure is greater than the saturation pressure for the given temperature). When water
and steam exist tagether at the saturation temperature, the steam is said to be “wet”.
The steam quality or dryness fraction is defined as the ratio of the mass of vapor to the
total mass. For example, in a water-vapor system, if the mass of the water-vapor i s 0.2
kg and the mass of liquid water is 0.8 kg, the quality or dryness fraction would be 0.2
or 20%. Accordingly, saturated water would have a quality or dryness fraction equal to
zero, while saturated steam would have a quality of one.

When the vapor is at a temperature greater than the saturation temperature, it is said to
exist as superheated vapor. In the superheated vapor region, the temperature and
pressure are independent properties.

Having produced the steam from the heat source or boiler it is then conveyed to the end
user where heat transfer takes place. The latent heat is given up and the steam
condenses, at the boiling point, to liquid, The resulting condensate still contains the
energy which was put in to bring the water to the boiling point. Therefore, good
practice ensures that the condensate is returned to the boiler for re-use.

If the pressure in the. system remains atmospheric, the steam will have a temperature
corresponding to that of boiling water at 1 atmosphere (0 atmospheres gauge pressure),
that is, 100°C. Under these conditions, the volume of the steam is considerable. T O
distribute the steam, large pipes would be needed, and some pressure drop will be
required to get any steam to flow to the point of use . Steam is therefore constrained in
the boiler and used at a higher pressure. This higher pressure is selecte d in accordance
with the needs of the process. The pressure reduces the specific volume of steam
(volume per unit mass) so that it can be distributed more easily in relatively small
pipework. This is shown clearly in the steam tables in Appendix A, where the third

Energy Conservation & Efficiency Project (ECEP)


 




Temperature

Exhibit 2.1
Typical Saturation
pressure-temperature Curve for water

 



SOME BASIC PRINCIPLES 2-3

column (“specific volume, ms 3/kg”) indicates how the specific volume falls with the rise
in pressure.

As the pressure rises, the boiling point of water (i.e. the saturation temperature) also
rises. This increases the usefulness of steam as a heating medium. However, the
amounts of heat required to raise water to the boiling point (sensible heat) and to
convert it to steam (latent heat) change with pressure, as shown in Exhibit 2.2. Since
the latent heat is usually regarded as the “useful” energy in the steam, high pressure
steam contains less useful energy (in terms of kJ/kg for example) at higher
temperatures. This also can be seen in the steam tables in Appendix A where the latent
heat is represented by the column of the enthalpy of evaporation, hfg.

Alternatives to steam for heat distribution have, of course, been developed. Circulating
hot water from a “hot water boiler “ (which does not boil) is the basis for many heating
systems. These systems are limited because the water temperature is quite low, unless
used with a pressurized system which is more costly. Higher temperatures can be
obtained using circulating hot oil systems. However, hot oil has the disadvantages of
poor heat content (low specific heat) and a poor heat transfer coefficient.

Some of the inherent advantages of steam as a means of conveying heat are as follows:

1. Because of its high latent heat, the high heat content of steam allows a given
amount of heat to be conveyed in relatively small [Link] results in a
lower first cost for installing a distribution system and reduced operating
costs for losses due to [Link] return lines are also relatively
small.

2. The weight of steam within a distribution network is relatively low

compared with the weight of hot water or hot oil in alternative heat
distribution systems. This simplifies the provision of piping supports or
hangers.

3. Steam flows in response to the pressure drop along the line and there is no
need to install expensive circulating pumps.

4. Steam provides a flexible system in which loads can be added and taken
away relatively easily. Variable speed circulating pumps, or differential
pressure control valves and bypasses, are avoided.

5. Steam has substantial advantages at the point of use. Heat transfer

coefficients from steam are often twice as high as those from water,
therefore, heat exchange equipment is smaller (and cheaper). Steam readily
fills spaces at uniform pressure, avoiding the expense of baffles, channels or
other flow distribution devices.

6. Steam is, in principle, odorless and non-toxic. Steam leaks do not provide
fire safety hazards and often do not require immediate shutdown of
processing equipment to effect repairs .

Energy Conservation & Efficiency Project (ECEP)


 




212 259 286 3O7 323 338 350 361 371 380 388

0 20 40 60 80 100 120 140 160 180 200

2800 1200

2600 1100

2400
1000

2200
900
2000

800
1500

700

600 Blu/lb
kJ;kq 1400

1200 500

1000
400

800
300

600
200
400
loo
200

100 120 134 144 152 159 165 170 175 180 184 188 192 195 198 Oc

Exhibit 2.2

Steam Properties

 



SOME BASIC PRINCWLES 2-5

2.2 STEAM TABLES

Steam tables are available which provide the properties of steam under different
conditions and are therefore indispensable for calculations involving both superheated
and saturated steam. For saturation conditions, knowing either the pressure or
temperature is sufficient to know the properties of saturated water and steam. TWO
tables are given in Appendix A for the saturated steam, the temperature and pressure
tables. In the temperature table, table A.1, the saturation properties can directly be
obtained with the knowledge of saturation temperature, while the pressure table, table
A.2, is used when the-saturation pressure of steam is known. The properties, which are
included in the saturated steam tables, are:

- temperature

- pressure

- specific volume of saturated water (v f) and saturated steam V g

internal energy of saturated wate r (uf), saturated steam (u g), and internal
energy of evaporation, ufg =ug- uf

enthalpy of saturted water (hf), saturated steam (hg), and enthalpy of

evaporation (latent heat), hfg = hg- hf

- entropy of saturated water (s f) , saturated steam (h g) , and entropy of

evaporation , sfg = sg - sf

Properties of wet steam (mixture of saturated water and steam) can be obtained from
the saturated steam tables if the quality or dryness fraction, x , of wet steam is known,
in addition of course of either the pressure or temperature, using the following formula

P wet = Pf+ x (Pg - Pf)

where Pwet ,, Pf and Pg are the required property of saturated water and saturated steam
at the given pressure or temperature, respectively .

For superheated steam it is necessary to known the pressure and temperature of the
system to obtain the values of the superheated steam properties. Table A.3 in Appendix
A provides the properties of superheated steam at various conditions.

Part of saturated steam systm is operated at 71 bar gauage supplied by a

boiler with efficiency of basis). Heat loss from the piping is to be 120,000

 



SOME BASIC PRINCIPLES 2-6

kJ per hour, and improved insulation Iis to be added which will reduce these losses by 60%. HO W
much fuel will be saved at the hoiler? Assume the gross heating value of the fue l IS 42,500 kj/kg
and the condensate IS retumel to the boiler at 92-C.

Calculation

Reduction in heat loss = 120,000 x 0.6 = 72,00 0 kJ/h

2. Using the saturated steam table (tabl e A.2):

Enthalpy of sat. steam at 7 bar = 2763. 5 kJ/kg
Enthalpy of sat. water at 7 bar = 697.2 2 kJ/kg
Enthalpy of evaporation = 2066.3 kj/kg
Note that 7 bar is approximately equal to 0. 7 Mpa

3. Steam saving: 72,000 kj/h 12066.3 kj/kg = 34.8 kg /h

.
4. Enthalpy of saturated water at 92 C = 385.3 kJ/kg

5. Energy needed to generate 34.8 kg/h steam at 7 bar

(2763.1-385.3) kJ/kg x 34.8 kg/h IO.80 = 103,434 kj/hr

6. Equivalent fuel savings

= 103.434 M/kg fuel 42500M/kg = 2.43 kg fuel oil per hour

In step No.6 the calorific value of oil is taken to be 42500 kJ/kg

2.3 MOLLIER DIAGRAM AND OTHER STEAM PROPERTY CHARTS

Some calculations, especially those involving the expansion of steam through turbines,
are most conveniently performed using a Moldier diagram. This diagram represents the
properties of steam graphically, as shown in Exhibi t 2.3. If the steam pressure and
temperature are known, then the condition of the steam can be determined from the
diagram. For example, steam at 100 0 kPa (note that 1 bar is equal to about 100 kPa,
o
so that 1000 kPa is equal to about 10 bar) and 250 C is a superheated steam because its
temperature is higher than the saturation temperature corresponding to the 1000 k Pa
o
(179.91 C). Using Moilier chart, the steam enthaipy and entropy are obtained to be
o
2965 kJ/kg and 6.9 kJ/kg C, respectively.

Steam properties can be represented in other convenient ways. Another common

method is the temperature-entropy (t-s) diagram as illustrated in Exhibit 2.4. The
temperature-entropy diagram can be used to represent steam cycles and to determine the
amount of work accomplished in cyclic processes.

 




Exhibit 2.3
Mollier Chart for Steam

 




Exhibit 2.4

 



3. ALL STEAM SYSTEM EFFICIENCY

3.1 SYSTEM COMPONENTS

fore discussing distribution systems in detail, it is important to understand the

principles involved in estimating the overall system efficiency. A typical steam system
will include the following elements:

1. boiler (or steam generator)

2. steam mains, including control valves, steam traps

3. condensate return system including piping, storage tanks and pumps

4. steam consuming equipment (the steam users).

The overall system efficiency relates the useful heat obtained by the user to the fuel
burned at the boiler and may be estimated from the individual efficiencies of each
element of the steam system. The following figures are illustrative for a 10 bar steam
system:

oiler Efficiency

Stack loss 16%

Structure radiation/convection loss 3%
Blowdown loss 2%
Boiler efficiency: ( 100 - 16 - 3 - 2) 79%

team Mains Efficiency

Efficiency, allowing for leaks and heat losses 92%

ercent of Steam Energy Available to the User

Assume that condensate is rejected at saturation temperature: from the steam

tables, the enthalpy of saturated steam at 10 bar is 2778. 1 kj/kg and of
condensate at the corresponding saturation temperature is 762.8 1 kj/kg. The
latent heat is 2015.3 kj/kg. The energy available is thus:

2015. 1 kj/kg / 2778.1 kJ/kg = 0.725 or 72.5%

ocess Equipment Heat Losses

Assume process equipment heat losses are approximately 3% . Efficiency is

thus 97%

verall System Efficiency:

0.79 x 0.92 x 0.725 x 0.97 = 0.5 1 = 51%

Energy Conservation & Efficiency Project (ECEP)


 



OVERALL STEAM SYSTEM EFFICIENCY 3-2

It is important to note that the overall efficiency is obtained by multiplying the

individual efficiencies. In the example quoted above, the efficiencies of each element
are reasonably typical for a factory, yet the overall system efficiency is hardly better
than 50%. This means that almost half the energy in the fuel burned in the boiler is
lost, and only half the energy is actually employed usefully in the process.

3.2 EFFECT OF IMPROVEMENTS

Small improvements in the efficiency of each element can result in a significant

improvement in the overall system efficiency. For example, improvements might be
made to the system described above through closer control of operating parameters,
repair of steam leaks, addition of insulation, recovery of condensate, etc., with the
following results:

Boiler Efficiency

Stack loss 14%

Structure loss 2%
Blowdown loss 1%
Boiler efficiency 83%

Steam Mains Efficiency

With reduced leaks and improved insulation 95%

User Efficiency

Percentage steam energy available, with half of condensate energy recovered

for boiler reuse: enthalpy of saturated steam at 10 bar is 2778.1 kJ/kg,
condensate 762.81 kj/kg. Assuming half the energy in the condensate is lost
(i.e., half is recovered), the enthalpy of condensate finally rejected would be
762.81/2 or 381.4 kj/kg. The energy available is:

(2778.1 - 381.4) / 2778.1 = 0.86 or 86%

Process Equipment Heat Losses

If losses are reduced to 2 % , efficiency is 98%

Overall System Efficiency with Condensate Return

0.83 x 0.95 x 0.86 x 0.98 = 0.66 = 66%

The fuel savings resulting from the combination of energy saving measures is
substantial:

( [ (1 / 0.51 ) - ( 1 / 0.66) ] / [ 1 / 0.51 ] ) x 100% = 23%


 



OVERALL STEAM SYSTEM EFFICIENCY 3-3

3.3 THE COST OF STEAM

As the discussion above demonstrated, there can be many losses in the steam system.
Hence, the cost of a tonne of steam depends on the magnitude of those losses. The cost
is also dependent on the temperature and pressure of the steam generated and on the
temperature of the feedwater used in its generation. The example below illustrates the
computation of unit steam costs.

Example

Saturated steam is generated at 10 har in a boiler firing mazout with a feedwater temperature of
.
85 C. The averall efficiency of the boiler is 75 %. The heating value of mazout IS 43 W/tonne.
If the price of mazout in Egypt is LE 50/tonne (and the international price of heavy fuel oil is
$120/tonne), what is the unit cost of steam?

Calculation

From the steam tables, the energy content of saturated steam (enthalpy) at 10 bar is 2778.1 kj/kg
.
steam; the energy content of the feedwaterat 85 C is 355.9 kJ/kg. Hence, the energy needed to
.
raise 10 bar steam from 85 C feedwater is 2778.1 - 355.9 = 2422.2 kj/kg steam or 2.42
kJ/tonne steam.

The overall boiler efficiency is 75%. Therefore, the amount of fuel that needs to be fired to
produce the energy input to the steam is 2.42 / 0.75 = 3.23 KJ fuel/tonne steam.

The heating value of mazout is 43 kj/tonne. The amount of mazout required to raise a tonne of
steam IS therefore 3.23 / 43 = 0.075 tonnes mazout/tonne steam. The price of mazout is LE 50
per tonne, so the cost of mazout per tonne of steam is 0.075 x 50 = LE 3.75 per tonne of steam.

At world oil prices of $120 per tonne, the cost of heavy fuel oil per tonne of steam would
computed the same way: 0.075 x $120 = $9 per tonne of steam .

 



4. HEAT TRANSFER FROM STEAM

4.1 RESISTANCES TO HEAT FLOW

The sole purpose of the steam generation and distribution system is to provide steam
which will give up its heat at the heat transfer surface. However, there are usually
several barriers to heat transfer at that surface, as illustrated in Exhibit 4.1.

The center section of Exhibit 4.1 represents the heating surface of any process in which
the steam does not come into direct contact with the material being heated. On the
right is the water which is being heated. There will be a layer of stagnant water and
possibly a film of scale on the metal surface, both of which impede heat transfer.

On the left of the metal wall is the steam. Usually there will be a film of scale built up
from rust and dirt from the equipment, There will also be impurities carried over from
the boiler into the steam. Removal of the scale films on both sides of the metal wall
must be done by regular cleaning.

Between the steam and the scale film there are two more films: condensate and air.
The condensate (like the water film on the other side of the metal) provides a major
resistance to heat transfer. Condensate is about 60 to 70 times more resistant than the
equivalent thickness of iron or steel and 500 to 600 times more resistant than copper.

The water film, or a similar film of the product which is being heated, may be kept to a
minimum thickness by mechanical stirring wherever possible. The condensate film,
formed as the steam condenses, must be allowed to drain rapidly from the surface. It
must be continually removed from the heat transfer equipment to prevent a liquid film
from building up.

In practice the water film is often thicker than it should be. This is due to the fact that
the steam itself is carrying an excessive amount of moisture. This wet steam is often
the result of carryover from the boiler. In this case boiler feed treatment compounds
may also be carried over. These will deposit on the heat transfer surfaces and further
impede heat transfer by increasing the scale film.

Apart from its effect on heat transfer, excess water at the heat transfer surfaces will
increase the load on steam traps and the condensate return system. It may increase the
load on the traps so that plant output is reduced.

The effect of the air film is even more serious. Air is more than 1500 times more
resistant to heat transfer, than iron or steel and 13,000 times more resistant than copper.
This means a film of air 0.025 mm thick will resist heat transfer as much as a wall of
copper 330 mm thick. The air film originates from air trapped in the system at start-up
and from air which is brought into the steam drum through feedwater (due to inefficient
operation of, or lack of deaerator). The air may also include noncondensible gases
produced in the boiler as a result of water treatment . If the air and noncondensibles are
not removed, heat transfer will be adversely affected by the insulating gas layer near
the metal surface. In addition, the temperature of a steam-and-air mixture will be

 




Exhibit 4.1

 



HEAT TRANSFER FROM STEAM 4-3

iower than that of pure steam at the same pressure. The result will be a reduction in the
heat transferred from the steam side to the water side.

The combined effect of some of these resistances to heat transfer is illustrated in Exhibit
4.2. In extreme cases, the air may accumulate in stagnant pockets in some parts of the
heat exchange equipment, resulting in cold spots in some places. Uneven surface
temperatures in drying cylinders are a good example. The provision of dry steam and
the complete removal of air are therefore essential if the highest heat transfer rates are
to be achieved.

4.2 WET STEAM

4.2.1 Causes of Wet Steam

On most boilers there is a certain amount of carryover of water which is mixed with the
steam as fine droplets. As pointed out in chapte r 2, the fraction of pure steam in the
mixture of steam and water is called the “dryness fraction”. For example, if the steam
contains 5% of water, the dryness fraction is 0.95. In a well operated, well maintained
and well designed system, a dryness fraction of 0.97 or better should be attainable.

Wet steam can be caused by inadequate drainage and poor insulation of the steam
distribution pipework. However, it is more commonly caused by the boiler, for various
reasons:

One problem stems from the limited size of many shell boilers. The
process of steam generation involves the formation of steam bubbles at the
heat transfer surface. These then rise to the water surface where they are
released into the steam space. The large amount of heating surface packed
into many boilers means considerable steam release at the wate r surface and
the surface becomes turbulent. The surface itself is high up in the boiler
because of the need to incorporate as many smoke tubes as possible. When
the steam bubbles are released, droplets of water are carried up int o the
steam space. In theory these should fall back. However, with a turbulent
surface close to the outlet connection it is inevitable that water is carried
over to give wet steam.

The above problem is aggravated by the way in which many boilers are
operated. The size of the steam bubbles is governed by the steam pressure
acting on the water surface. A shell boiler may be designed to operate at
around 14 bar. If it is operated at a lower pressure, say 7 bar, then the size
of the steam bubble is increased, as is the risk of carryover. Eve n when
operated correctly, pressures can fall as the boiler attempts to cope with a
sudden peak. This explains why carryover and wet steam occur at times of
high demand.

Another factor causing wet steam can be improper water treatment. Some
dosing can cause foaming of the boiler water, especially when chemicals are

 




Exhibit 4.2
Typical Steam Heat Transfer Profile

 



HEAT TRANSFER FROM STEAM 4-5

dumped into the feed tank in a mass instead of being supplied at a steady
rate. Not only does this foaming give rise to water carryover and wet
steam, but the treatment compound itself is carried over. It is therefore
unable to condition the boiler water as intended and merely fouls up
strainers, valves or heat transfer surfaces.

Within the steam distribution system itself, there are various potential problem areas
which can contribute to making wet steam . Indeed, in severe cases, slugs of water can
be found in steam pipes. Moving at steam velocity, these can cause severe damage to
pipework and fittings.

The formation of wet steam begins as soon as the steam leaves the boiler. Some of it
starts to condense as heat losses take place from the pipework. The rate of
condensation will be particularly heavy at start-up when th e pipework is cold.
Condensation will continue under operating conditions, however well insulated the
pipework may be.

Exhibit 4.3 is a diagrammatic representation of the way in which droplets of condensate

can build up in a straight pipe. Ultimately they can form a solid slug and cause
destructive water hammer.

The creation of water hammer will be accelerated if th e pipework is allowed to sag.

This can occur when pipe hangers fail or when support is inadequate. There is a
similar danger when water is allowed to collect at low points in the pipework.

Exhibit 4.4 shows an example where th e pipework is half flooded. This means quite
simply that steam velocity will be doubled in the remaining free area of the pipe. The
chance of slugs of water being picked up to create water hammer conditions is also
increased. The restricted free area will also mean high pressure drops even though the
pipework was originally sized correctly.

4.2.2 Alignment and Drainage of Steam Mains

Proper alignment and drainage of steam lines means observing a few simple rules:

1. Steam lines should be arranged with a fall in the direction of flow. A fall
of around 40 mm in 10 m is generally adequate. This ensures that
condensate is carried easily by the steam flow towards the next drain point.
At shut down, it ensures complete drainage of residual condensate. If
horizontal pipework is absolutely necessary due to site conditions or to
reversible flow, then more frequent drainage will be necessary.

2. Saturated steam lines should be drained at regular intervals. The distance

between drain points will depend on line site, location and the frequency of
start-up, but intervals of 30 to 50 m are usual. Lines should also be drained
at any low point where condensate can collect.

 




Steam

Steam

Exhibit 4.3
Condensate Buildup in Steam Piping

Steam
I/ j
i1
==>

Exhibit 4.4
Condensate Accumulation
Reduces Steam Flow

 




HEAT TRANSFER FROM STEAM ______________

4-7

3. Drain points are most effective wher e pipework hanges direction. Long
lengths of pipework over horizontal ground call for relay points as shown in

Exhibit 4.5. These make ideal drain points since the change of direction
helps to separate out entrained droplets of moisture.

4. When drainage has to be provided in a straight lengt h of pipe, a large bore

pocket should be provided, as shown in Exhibit 4.6. Droplets of water are
generally carried along the pipe wall and these will be carried past a small
bore tapping. Full bore pockets are ideal and are practical i n pipework up
to 100 mm diameter. For larger steam mains, the pocket can be 2 or 3
sizes smaller than the line being drained.

5. Pipework should be arranged so that places where water can collect are
avoided. Exhibit 4.7 shows th e incorrect use of a concentric reducer which
allows condensate to collect. The correctly installed eccentric reducer will
ensure that condensate ca n flow to the next dram point. Similar problems
can occur with some valves and fittings. The strainer shown in Exhibit 4.8
is a potential cause of water hammer. In this case, much of the screening
area is ineffective, being out of the steam flow . Strainers above 25 mm in
size should be fitted on their side to avoid these problems.

6. Globe valves having under and over construction can also form a weir and
prevent condensate from flowing to the next drain point. This happens
when they are fitted in “normal” position in a horizontal line. A gain, this
can be avoided by fitting the valve on its side.

7. Branch connections should always be taken from the top of an y main so that
the driest possible steam is supplied. A connection from the bottom is the
worst possible arrangement. Such a connection wbuld act as a drain pocket
and the result would be a very wet steam supply.

8. Where a branch line drops from a high level main to serve a plant at floor
level then the low point before reaching any equipment should also be
drained, as shown in-Exhibit 4.9 This is particularly important if a control
valve or reducing valve is fitted in this line: a valve designed for steam
cannot work efficiently in a pool of water.

9. At the point of steam usage, the use of a separator should be considered.

This provides an adequate drain point for any droplets of moisture which
are being dragged along the pipe wall. In addition, the slower velocity
through the separator body and the arrangement of the baffles will ensure
that droplets of water carried along in the main steam flow will also be
separated out. This guarantees the supply of dry steam, particularly
important for highly rated plants, or for equipment like sterilizers where
steam comes into contact with the load o r product. Two kinds of baffle
type separators are shown in Exhibit 4.10. Separators are discussed further
in Section 4.3.

Energy Conservation & Efficiency Project (ECEP)


 




Condensate

Exhibit 4.5
Proper Trapping of Long Steam Mains

 




Steam

Exhibit 4.6
Proper Drainage of Steam Piping

Correct

incorrect

Exhibit 4.7
Proper Reduction of Steam Piping

 




S team

Exhibit 4.8
Condensate Accumulation in
Vertically Mounted Strainer

Steam
Trap

Condensate

Exhibit 4.9
Drain Low Points of Steam Supply
to Equipment

 




Exhibit 4.10
BaffIe-Type Steam Separator

 



HEAT T 4-12

10. use there is always a risk of water hammer in steam mains, steam traps
fairly robust. The thermodynamic trap shown in Exhibit 4.11 is a
i ce,as is the inverted bucket steam trap in Exhibit 4.12. If the
to be fitted in an exposed position, it is useful to fit the
namic trap with an insulating cover to slow down its rate of
,while the bucket trap should always be properly insulated.

11. If the lines are large or are subject to frequent start-up, additional air
venting s provided. A good location is the end of any steam line,
as shown in Exhibit 4.13.

most rules governing steam systems, these comments provide general

only; there will always be exceptions. A case in point is provided by the
problem of a a across rising Exhibit 4.14 shows one
tions made in 3 above. However, this
and it may be to run a line with a rise
rather than a fall in the direction of flow.

4.3.1 nci ples

A se r is desig to remove entrained water particles from steam so that they can
bed away as densate through a conventional steam trap. This is done using
change of direction d change of velocity.

Exhibit 4. 10 shows simple but effective baffle type separators. In each case, the body
has a cross sectional area many times greater than that of the approach pipework. This
means that there is a sudden drop in steam velocity. Water droplets have a somewhat
greater weight and momentum. If they fail to drop out due to the velocity change, they
impinge on the baffles and are guided down to the drain point.

This t of separator has the advantage that it can also serve as a drain point for
condensate which has already separated from the steam and is running along the bottom
of the steam main .

4.3.2 Separator Efficiency

Claims are sometimes made about the efficiency of separators . A generally high figure
may be quoted suggesting that almost all free water is removed. Such figures are
misleading on two counts.

In the first place, the performance of the separator varies the conditions. Total
moisture content, droplet size and particularly steam velocity will all have a bearing on
any results .

 




Exhibit 4.11
Thermodynamic Steam Trap

Exhibit 4.12
inverted Bucket Steam Trap

 




Exhibit 4.13
Air Vent at the End of Steam Main

Exhibit 4.14
Steam Piping Over Rising Ground

 



[Link] TRANSFER FROM STEAM 4-15

Secondly, and more importantly, it is not even possible to measure separator efficiency.
Although calorimeters are available to give a measurement of dryness fraction, it is
almost impossible to take a representative sample from the steam pipe either upstream
or downstream of a [Link] at the center of a pipe will travel at the highest
velocity and will be relatively free of water droplets. Towards the pipe wall, the
velocity will be lower, and the steam flow will carry more moisture. The pipe wall
itself may be covered by a film of water being dragged along at some lower velocity.
Results will depend on the point from which any sample is taken. In practice, it is
therefore impossible to get any meaningful “before” and “after” figures on which to
base any claim for separator efficiency.

4.3.3 Some Separator Applications

Separators provide the best possible way of removing entrained moisture from the
steam [Link] on the layout of the steam distribution system, their use may
be highly desirable or even [Link] is not possible to list all possible applications
but it is worth looking at some typical installations.

Because many wet steam problems arise from the boiler, separators are commonly fitted
fairly soon after the boiler crown valve. They remove much of the water carried over
when priming takes place, and ensure a much drier steam supply to the entire system.
Priming will often carry over a quantity of boiler feed water treatment compounds.
Although not designed for that particular purpose, the separator helps to prevent these
contaminants being carried further into the steam system. However, the contaminants
can cause probiems with the separator trapping arrangements, which should be selected
with this problem in [Link] 4.15 shows a separator where the strainer preceding
the trap has been fitted with a blowdown cock to facilitate frequent cleaning.

Separators are also used to protect valves or fittings which can be damaged by the
presence of wet [Link] reducing valves often benefit from the provision of a
drained separator on the upstream side (Exhibit 4.16) .

Another very obvious requirement for separators is in applications where steam comes
into contact with product. Water in the steam supply to bread ovens can result in
unwanted marking of the loaves;similar spotting can occur in some dye machines. In
hospital sterilizers the result may be more serious, with wet packs or incomplete
sterilization of the load space. Exhibit 4.17 shows the provision of dry steam to a
spotting gun as used in the dry cleaning industry while the same arrangement is also
relevant to the hand steaming iron used in some laundries.

Although the applications just described are fairly obvious in that excess water and its
effects are clearly visible, there are other applications where the hidden benefits may be
just as [Link] heat transfer surfaces will benefit if excess water in steam is not
allowed to spray onto heat transfer surfaces. For example, in the laundry industry the
full output of the steam heated bed of a flatwork ironer may be crucial if design
performance is to be [Link] ironers have achieved this performance only after
a separator was installed in the steam supply.

Energy Conservation & Efficiency Project (ECEP)


 




Separator

Steam Trap V

Exhibit 4.15

Steam Separator with Blowdown Cock

for Easy Cleaning

Exhibit 4.16

Protection Control Valve With

Upstream Separator and Drain

 




Exhibit 4.17
Separator Installed to Ensure
Dry Steam to Spotting Gun

 



HEAT SFER FROM STEAM 4-18

4.4 AIR IN STEAM SYSTEMS

4.4.1 Sources of Air

It is fairly evident that when a steam plant is about to be commissioned the steam space
will be full of air. This must be removed before the steam space can become fully
effective. What is not always realized is that this situation occurs every time the plant
is shut down. Residual steam will condense and an initial vacuum may be formed,
drawing air in to the steam space. Air can be drawn through steam traps, air vents,
valve glands and other minute leaks which are otherwise unnoticed. Again, it must be
removed before the steam space becomes effective.

Air can also enter the system via the feed water. All raw water supplies contain
oxygen, introduced into the feed tank with make up water. Air also contains carbon
dioxide: although the quantity is small (typically around 0.03%) its solubility in water
is some 30 times that of oxygen. Carbon dioxide is therefore released during steam
formation along with oxygen.

The release of these unwanted gases is aggravated by the fact that treatment to remove
some salts causing hardness will result in the production of further quantities of carbon
dioxide. Any undissolved gases will have an adverse effect on heat transfer if retained
in the steam space.

It is worth considering the many separate effects of air in the steam system. All tend to
reduce output and increase steam consumption and are considered in the following
sections.

4.4.2 Effects of Air

[Link] Delay at Start-Up

The problems of air are most obvious, but often unrecognized, at [Link] is
turned on and there then follows a period of “warm up” as the equipment, plant or
space comes up to temperature. Steam may be turned on up to 2 hours early so that
equipment may be warmed up by the time the shift starts. It may be turned on to a
jacketed pan for 20 minutes because experience shows that this is the time required to
boil the contents.

In both cases a significant part of the warming up period is due to the time taken to
remove initial [Link] is evident that until air is removed, steam cannot enter the piping
or jacket. Inadequate provision for removing air means extended warming up times
with increased standing losses.

Conservation & Efficiency Project (ECEP)


 



HEAT TRANSFER FROM STEAM 4-19

[Link] The Steam/Air Mixture

Air in the steam space is commonly removed through the steam [Link] that the
steam space has a relatively small cross-sectional area the air is simply displaced by the
steam as shown in Exhibit [Link] steam at 7 bar the pipe or cylinder will have a
o
surface temperature of 170.5 C. If the mixture contains a mere 10 % of air, the
o
temperature fails to 166 [Link] the temperature drop may appear small, it will
nevertheless mean reduced performance and can be critical where a specified pressure
and temperature are required to enable curing or some such process to take place .

[Link] Resistant Films

As already indicated, the effect of air on the temperature of a steam and air mixture is
comparatively smaIl (although perhaps crucial in certain cases). While it might be
thought that 10% air in the mixture is an unreasonably high estimate, the problem is
that air will collect in certain places, and its effect is out of all proportion to the average
quantity involved.

In the normal course of events, steam comes into contact with the heat transfer surface
and condenses as it gives up its heat. The condensate drains to the low point in the
steam space to be replaced by more steam. The steam and air mixture at its slightly
lower temperature behave in roughly the same [Link] the steam condenses and
runs away, the air tends to remain at the heat transfer surface where it forms a [Link]
mentioned previously, air is a particularly poor conductor of heat.

The significance of this may be better understoo d by reference to figures. Thermal

o
conductivity expressed as -watts per square meter per mete r thikness for 1 C
o
temperature difference (W/m2 C) is 0.026 for air, 0.6 for water, 46.0 for iron and
340.0 for copper.

This means that air is more than 12,000 times more resistant to heat transfer than
copper, or 1,720 times more resistant than iron. However thin the air film may be, it
will have a significant effect on heat transfer and will reduce plant output.

[Link] Cold Spots And Uneven Temperatures

Previous sections have indicated the way in which air lowers the effective steam
temperature and provides a barrier to heat transfer. These effects may go unnoticed but
the accumulation of air is generally more obvious. In the majority of process
equipment, when the steam is initially turned on, it pushes the air ahead of it so that the
air collects at some point remote from the steam inlet (Exhibits 4.18, 4.19). These
pockets of air are relatively cold so there is immediately a loss of heat output. More
seriously, air pockets result in uneven temperatures at the heat transfer surface. This
can cause wet spots in the output from cylinder drying machines or can interfere wit h
the smooth flow of sheets from a laundry calender . In multi-tube heater batteries these
cold spots mean unequal expansion of the tubes and the resultant stresses can be the
cause of joint failures .

Energy Conservation & Efficiency Project (ECEP)


 




Exhibit 4.18

Displacement of Air by Steam in Piping

Exhibit 4.19
Air Accumulation in Heat Exchanger Tubes

 



HEAT TRANSFER FROM STEAM 4-21

4.4.3 Air Removal

The removal of air is essential and can be- carried out either by manual air venting or
automatic air venting. Manual air venting has the disadvantage of relying upon the
operator knowing just when and how often the vent should be opened and, more
difficult still, knowing when to shut it again (especially since an air/steam mixture
looks exactly like steam).The better choice is obviously an automatic air vent.

It is not sufficient, however, just to make provision for air ventin g as such the rate of
air removal must be [Link] is during the stage when steam is turned on to the
plant that provision must be made for the rapid removal of air, The more rapidly it is
removed, the less will be the possibility of air being diffused and mixed with th e steam ,
and the more rapidly will the steam space be fully occupied with steam, and the less
time will it take for the plant to warm up to production temperature.

Air and steam are at different temperatures, a fact which is used in the design of the
common thermostatic air vent. The air vent will discharge cold air and steam/air
mixtures, closing when the temperature of the mixture approache s that of pure steam.
A typical thermostatic air vent is shown in Exhibit 4.20.

Air vents usually discharge to atmosphere so that they are not restricted in any way.
However, there can be problems as they operate, and there is always the chance of
some escape of steam after they have [Link] is therefore essential that the discharge
from any air vent is piped to a safe place.

Air vents are required in three different situations. They may be used to:

1. Augmentair venting atthe condensate drainagesteam trapping) point,

mainly toensure quickerwarming up

2. Handle air from large steam spaces in order to avoid forming and air
mixtures

3. Handle air removal from steam spaces where cold pockets of air can form

These three requirements all overlap. The large steam space can suffer from cold spots
while the problems of resistant films are ever [Link] to fit air vents can be
considered by reference to the following diagrams showing typical applications.

Exhibit 4.21a shows an air vent fitted in parallel with an inverted bucket trap which is
relatively slow to vent air This arrangement might be used on a steam main which is
subject to frequent start-up. It will be noted that the air vent discharges into the
condensate line so it will be preferable if this line drains by gravity.

An alternative layout for the end of 3 steam main is shown in Exhibit 4.2 1 [Link]
arrangement is particularly suitable for large steam mains. It can be seen that the air
vent discharges open ended.

Energy Conservation & Efficiency Project (ECEP)


 




Exhibit 4.20
Thermostatic (Balanced Pressure) Air Vent

 




(a) Discharge to Condensate

Discharge to Ambient

Exhibit 4.21
Installation of Air Vents in Parallel
with Steam Traps

 



HEAT TRANSFER FROM STEAM 4-24

In the case of a large steam space, the positioning of any air vent is dictated by the way
in which steam is [Link] Exhibit 4.22a the steam is injected at the bottom and the
air is displaced upwards. Exhibit 4.22b shows the alternative of downward
displacement: with steam fed from the top, the air vents are at low level.

In both cases, each air venting point is fitted with its own automatic air vent. The
“grouping” of a number of drainage points to a single steam trap is generally not
recommended because steam from one point reaching the trap will cause it to close
before drainage from the other points is complete. Exactly the same thing can occur
when a number of air venting points are grouped to a single automatic air vent.

Exhibit 4.23 shows a jacketed pan and is an example of an awkward steam space.
Steam admitted to the jacket will condense and drain through the steam trap while air
will be trapped on the side remote from the steam inlet. Looking at the product
surface, one would observe vigorous boiling in line with the steam inlet and little or no
reaction on the opposite side. An air vent fitted in the position shown will remove the
air pocket and ensure optimum performance.

Energy Conservation & Efficiency Project (ECEP)


 




(a) Bottom Steam Injection, Air Vents High

Top Steam Injection, Air Vents Low

Exhibit 4.22
Air Vent Installation Relative to Steam Supply

 




am

Air Discharge

Float Type Steam Trap with

Thermostatic Air Vent

Air Vent Capability Both in the Steam Trap and Jacket

Exhibit 4.23
Jacketed Pan

 



5. SIZING OF STEAM DISTRIBUTION SYSTEMS

This section provides guidance on pipe sizing and other aspects of steam distribution
system design. A poorly designed system is not likely to be energy efficient. It can
also cause problems of getting the right quantity of steam to the user, thereby causing
further energy inefficiencies.

5.1 SELECTION OF PRESSURE

The pressure of a steam system depends primarily on the process temperature

requirements. Low pressure steam has a higher latent heat per unit mass then high
pressure steam and is therefore a better heat carrier. However, low pressure steam is at
a lower temperature and therefore its use will result in larger heat transfer surfaces for a
given heat duty. Low pressure steam also occupies a large volume and distribution
requires large pipes. The pressure of a system is therefore the result of a balancing of
all the factors involved.

5.2 PIPE SIZING

The sizing of steam distribution pipes is clearly linked to the system pressure.
Although it is sometimes necessary to design a system with futur e loads unknown,
extreme oversizing should be avoided for two reasons:

1. Oversized piping is more expensive, and the cost of insulation , supports and
fittings must also be taken into account.

2. Oversized piping has higher heat losses (for comparable insulation) because
of the greater surface area per meter run. This in turn leads to a greater
steam condensation and the need for a larger condensate return system.

The problems of undersized pipework are more obvious. The lines are unable to supply
steam at the required rate and steam starvation at the using end is indicated by reduced
pressure. The result is a reduction in output from the heaters or process plant. A t the
same time, the high steam velocities can cause problems due to erosion and increas e the
possibility of water hammer.

There are two methods of sizing pipes, both of which have an unknown factor which
must be assumed:

1) sizing based on velocity

2) sizing based on pressure drop.

If we assume a velocity, our calculations are based on the specific volume of th e steam
being carried in relation to the cross sectional area of the pipe. Practical experience
shows that reasonable velocities for dry saturated steam mains are 25-35 m/s. These
should be regarded as maximum values above which noise and erosion will take place,
particularly if the steam is wet.

Energy Conservation & Efficiency Project (ECEP)


 



SIZING OF STEAM DIS BUTION SYSTEMS 5-2

The carrying capacities of pipes for various steam velocities are shown in Appendix B
(Table 1). The table may be used to size short branch connections based on the
velocity range indicated above. However, the resultant pressure drops can be high, so
longer lengths of pipe should be installed giving a steam velocity of no more than 15
m/s.

This method is clearly easy and convenient but it provides no guarantee of achieving the
desired pressure at the using end. Where pressures (and therefore temperatures) are
important or where distances are long, it is better to size on pressure drop. Having
decided the final pressure required, the pipe size is calculated using a formula which
can be simplified to:

(P1-P2) /L = F

Where:
PI is a factor based on the initial pressure
P2 is a factor based on the final pressure
L is the length of travel (including a n allowance for
valves and fittings)
F is the pressure drop factor.

The factors P and P2 can be determined from Appendix B, Table 2. Table 3 shows
how the calcul ated pressure drop factor can be used to determine the carrying capacity
of various sizes of pipe. This same table provides information on velocity so that this
can also be checked. A worked example is included as Appendix C.

5.3 EXPANSION

A particular problem with steam mains arises from the expansion of the system at start-
up. The problem is illustrated in Exhibit 5.1, which shows the expansion of steel pipes
o
at different temperatures when installed a t l16 C. Wherever there are long runs of
straight pipe, some provision must be made for expansion.

Exhibit 5.1

Expansion of Hot Piping with Temperature

Final Temperature Expansion per 30 m

66 C
o
19 mm
93 29
121 41
149 50
177 61
204 74
232 84
260 97

 



SIZING OF STEAM DIS UTION SYSTEMS 5-3

It is common to use expansionfittings in conjunction with fixed anchor points.

fittings include:

1. Full loop: The loop should be fitted in a hori with the

downstream side passing below the upstream side (Exhibit 5.2).

2. Horseshoe loop:
(Exhibit 5.3).

3. Sliding joint:This is soemetimes used because it s up little room, but

the pipeline must be rigidly salignment can cause
the pipe to bend. The glan ires regular maintenance (Exhibit
5.4).

4. Bellows: Also occupying little tting has the advantage

of having no packing (Exhibit5.
Exhibit 5.5 st be rigidly anchored.

 




Exhibit 5.2
Full Loop Pipe Expansion Fitting

Exhibit 5.3

Horseshoe Loop Pipe Expansion Fitting


 




Exhibit 5.4
Sliding Joint Pipe Expansion Fitting

Exhibit 5.5

Bellows Pipe Expansion Fitting


 



6. STEAM TRAPS

6.1 THE FUNCTIONS OF STEAM TRAPS

It is extremely important that steam distribution systems always be properly drained.

Steam traps are the key to optimum steam/condensate system operation, and as such,
have three basic functions:

1. Remove condensate: condensate must be allowed to pass through the trap

quickly and completely, to maximum heat transfer.
2. Remove air and other noncondensible gases: air and gases in the system
reduce heat transfer area, and lower the overall heat transfer coefficient. In
addition, 02 and CO2 are corrosive in the presence of condensate.

3. Prevent steam loss: the trap must minimize live steam loss while it passes
condensate, air and noncondensibles.

6.2 TYPES OF STEAM TRAP

There are three basic types of steam traps, operating on three basic physical principles
to distinguish between steam and condensate :

1. Thermostatic (temperature difference): The thermostatic trap senses

temperature, and operates on a temperature difference between steam and
condensate after being subcooled a little by the valve body. In addition, the
steam has a higher heat transfer coefficient. This can have an immediate
effect on the thermostatic sensor. The sensor cools in the presence of
condensate, and opens a valve to release it. Upon contact with live steam
the valve is closed to prevent steam loss .

2. Mechanical (density difference): This trap relies on the difference in

denisty between steam and condensate for its operation. A float or bucket
rises in the presence of condensate, and falls when only steam is present.
This mechanical movement causes a valve to open in the presence of
condensate, and close in the presence of steam.

3. Thermodynamic (change of state): This third type of trap encompasses a

group of apparently unrelated devices. These include thermodynamic,
impulse and labyrinth traps, and even the simple orifice plate. All rely on
the fact that hot condensate, released to a lower pressure, can “flash off” to
give a mixture of steam and water.

6.2.1 Thermostatic Traps

There are several subcategories of thermostatic traps, as described in the subsections

which follow.

Eergy Conservation & Efficiency Project (ECEP)


 




STEAM TRAPS 6-2

[Link] Liquid Expansion

The liquid expansion trap shown in Exhibit 6.1 is one of the simplest thermostatic
traps. An oil-filled element expands when heated to close the valve against the seat.
An adjustment allows the element to be moved relative to the seat. This effectively
alters the temperature at which the valve opens, permitting regulation of the
temperature of the trap discharge.

A problem is that the temperature of steam varies with pressure. The trap, on the other
hand, can only be set to operate at a fixed temperature. Exhibit 6.2 shows the
saturation curve for steam together with the response line of the liquid expansion trap to
illustrate the problem. At steam pressure Pl the condensate would have to cool by only
a small amount and trapping would be acceptable. However, if steam pressure is
increased to P 2 then condensate has to cool appreciably before passing through the trap.
Since the condensate continues to accumulate as it cools, serious waterlogging will take
place. At a reduced steam pressure P3 the trap will blow live steam.

Although this trap provides a useful device for discharging condensate at a fixed
temperature, it can only be used in those applications where significant waterlogging
can be tolerated.

[Link] Balanced Pressure

An improvement is provided by the balanced pressure trap shown in Exhibit 6.3. As

the name suggests, this trap is “balanced” to cope with pressure (and therefore
temperature) variations. The oil filling is replaced by an evacuated element containing
a small quantity of a liquid with a boiling point somewhat lower than water.

In the cold conditions which exist at start-up, the element is contracted. Th e valve is
off its seat and is wide open, allowing the prompt removal of air. This is a feature of
all thermostatic traps and explains why they are also used as air vents.

As condensate passes through the trap, heat is transmitted to the liquid in the element.
Before steam reaches the trap, this liquid boils. The vapor pressure within the element
causes it to expand and the trap shuts. After a while, heat loss from the trap cools the
condensate surrounding the element and the filling condenses. The element contracts to
open the valve and releases condensate until it again approaches steam temperature, and
the cycle is repeated.

The temperature below steam temperature at which the trap opens is governed by the
filling. At the same time the fact that the element is thin-walled means that the boiling
effectively takes place at the operating pressure . The result is the response line shown
in Exhibit 6.4.

A problem here is the thin-walled element, which is sometimes susceptible to damage

due to water hammer or corrosion . This can be countered by the use of stainless steel,
while other designs use a completely filled element. Exhibit 6.5 shows a modem
element with considerable resistance to damage by water hammer or corrosion.

Energy Conservation & Efficiency Project (ECEP)


 




Scar

Exhibit 6.1
Liquid Expansion Steam Trap

Exhibit 6.2

Response of Liquid Expansion Steam Tra p


 




Exhibit 6.3
Balanced Pressure Steam Trap

Exhibit 6.4

Response of Balanced Pressure Steam Trap


 




Exhibit 6.5
Improved Balanced Pressure Element

 




6.2.1. c

The most robust thermostatic traps are those using bimetal strips of dissimilar metals
6.6 shows a simple
Exhibit 6.6
will vent air. As
bimetal deflects so
ds on the
trap, an

tion,bimetallic steam traps co

Exhibit 6.7 Condensate passing through
the seat while line pressurewil
condensate increases, there is an
force. The result is the
on Exhibit 6.2 but is still

Use is therefore made of different types of bimetal strips in a single stack to get the
response line to change as shown in Exhibit 6.8(ii). One set of bimetal deflects to give
the response A At higher temperature a second set of bimetal strips contributes to
give response At still higher temperature, the third set of bimetal deflects to
produce [Link]’purpose is to follow the saturation curve as nearly as possible.

The trap response line can be effectively moved up or down by adjusting the position of
the valve relative to the seat, and there is a temptation in some cases to set the trap
close to the saturation line. This is unwise on several counts. In the first place,
bimetal suffers from hysteresis, as shown in Exhibit 6.8 (iii). Bimetal will deflect and
close avalve ng water is at a somewhat higher temperature than the
bimetal. It n when the surrounding water has cooled to a lower
temperatur e There is therefore a difference between closing and
opening temperatures and there is no question of setting the trap at a precise
temperature.

Similarly, the tting depends on a balance between the closing force provided by the
bimetal and an opening force provided by line pressure. Clearly any back pressure
acting on the downstream side of the valve can upset this [Link] many
condensate return systems do impose a back pressure which varies with load conditions,
so the trap setting will [Link] if this type of trap is set to operate close to
steam temperature, there is the danger that under certain conditions it will blow live
steam as shown in Exhibit 6.8 (iv). Furthermore, operating around steam temperature
can produce wiredrawing of the valve and seat so that they are incapable of complete
closure. Bimetallic traps are therefore adjusted to ensure that they will hold back
condensate at all times.

esigns use shapes, stacked in pairs, to pull the valve onto the seat
as shown in Exhibit 6.9. One particularly successful shape is shown in Exhibit 6.10

Energy Conservation & Efficiency Project (ECEP)


 




Exhibit 6.6
Simple Bimetal Steam Trap

 




Exhibit 6.7
Bimetal Trap with Downstream Valve

 




(ii)

(iii) Valve Closed

Exhibit 6.8
Response of Bimetal Steam Trap

 




Exhibit 6.9
Operation of Bimetal Steam Trap

 




Exhibit 6.10
Shape of Spirax l Element

 



STEAM TRAPS 6-12

which indicates the way in the three legs of different length operation
at different temperatures.

Unfortunately, as the bimetallic strips hysteresis can cause them to take up

something of a permanent set.

6.2.2 Mechanical Traps

These traps may be divided into two basic categories: ball float traps, and bucket float
traps.

[Link] Ball Float

Like other mechanical traps, the ball float type senses the difference in density between
steam and condensate. In the case of the trap shown in Exhibit 6.11, condensate
reaching the trap will cause the ball float to rise, lift the valve off its seat and release
[Link] release of condensate from the trap is automatically modulated with its
[Link] the float rises due to increased condensate flow, the valve opens wider.

The problem is that air cannot pass through the flooded valve. Early traps of this kind
were vented using a manually operated vent on the top of the body. Modem traps use a
thermostatic air vent as shown in Exhibit [Link] air vent allows initial air to pass
while the trap is also handling condensate. The air vent is virtually the same as the
element of a thermostatic steam trap and is situated in the steam space above the
condensate level. After purging the air, contact with steam will shut the thermostatic
vent; it will generally remain shut until air accumulates again, displacing the steam and
causing the vent to open.

In many ways the float trap with thermostatic air vent is closest to the ideal steam trap .
It will handle air easily and will get rid of condensate just as soon as it is [Link]
will do this regardless of changes in steam pressure. Unfortunately the trap is
comparatively large, and can be expensive. A second disadvantage is that the ball float
and valve mechanism is sensitive to damage by water hammer.

[Link] Inverted Bucket

The inverted bucket trap is shown diagrammatically in Exhibit 6.13. As the name
implies, the working portion consists of an inverted bucket attached through a lever to a
[Link] essential part of the trap is the small air vent hole in the top of the bucket.

Exhibit 6.14 shows the method of operation. In (i) the bucket hangs down, pulling the
valve off its [Link] flows under the bottom edge of the bucket, filling the
body and flowing away through the outlet. When steam reaches the trap it collects in
the top of the bucket making it [Link] bucket then rises as shown in (ii) and
closes the valve. The trap remains shut until the steam in the bucket is dissipated. This
results mainly from condensation due to heat loss from the body, while some steam

 




Exhibit 6.11
Float Trap

Exhibit 6.12
Float Trap with Thermostatic Air Vent

 




Exhibit 6.13
Inverted Bucket Trap

Exhibit 6.14
Operation of Inverted Bucket Trap

 



STEAM TRAPS 6-15

bubbles out of the bucket through the vent hole . In (iii) the bucket is about to lose its
buoyancy. It then sinks, pulling the main valve off its seat. Accumulated condensate is
released and the cycle is repeated.

Air reaching the trap at start-up will also give the bucket buoyancy causing the valve to
close. Air does not condense, so the air vent hole is essential to allow air to escape out
of the bucket into the top of the trap . The hole is small to prevent excessive steam loss,
but as a result, the trap is relatively slow at venting air. Some steam loss does occur
through this hole, and the trap may give a dribbling action under low load conditions.

Because steam and condensate are discharged into the inside of a cylindrical bucket, the
trap is generally more robust than the ball float type. It is also relatively inexpensive.
On the other hand, the trap is poor at venting air, and its intermittent operation means
that the flow of condensate is not always continuous.

6.2.3 Thermodynamic and Miscellaneous Traps

Several other types of trap are available, including thermodynamic traps which are
probably the most common type used in industry.

[Link] Thermodynamic

This is by far the most commonly used trap which is based on the principle that hot
condensate will produce flash steam when the pressure is reduced. The trap is
supremely simple (Exhibit 6.15), consisting of a body with a flat working face having a
center inlet and annular groove connected to the outlet. The only moving part is the
disc moving inside the control chamber or cap.

Exhibit 6.16 helps to explain how the trap operates. In (i) disc A is raised from the
seat C by incoming pressure, allowing air and condensate to pass radially outwards
under the disc before discharging through outlet B. As the flowing condensate under
pressure approaches steam temperature, it experiences a drop in pressure through the
inlet orifice, and some of the condensate flashes to steam. This means that the rate of
fluid flow radially outwards under the disc is greatly increased, and accordingly the
pressure is reduced and the disc is therefore drawn towards the seat.

This reduced pressure alone is not sufficient to cause the disc to seat. However, further
flash steam flows around the end of the disc resulting in a pressure build-up in the
control chamber D as, shown in (ii). When the steam pressure in chamber D acting
over the full area of the disc exceeds the incoming condensate pressure acting on the
much smaller inlet area, the disc snaps shut covering the inlet orifice. This snap action
is important. It removes any possibility of wiredrawing the seat, ensuring no leakage.

This state is shown in (iii) and will persist until condensation in the control chamber
reduces the pressure over the disc. As shown in (iv), the incoming pressure will
eventually exceed the control chamber pressure and the disc will be raised, starting the
cycle all over again.

Energy Conservation & Efficiency Project (ECEP)


 




Exhibit 6.15
Thermodynamic Steam Trap with
Integrated Strainer

 




(iv
(iii)

Exhibit 6.16
Operation of Thermodynamic Steam Trap

 



STEAM TRAPS 6-18

The rate of opening clearly depends on steam temperature and also on ambient
conditions. Most traps will stay closed for between 20 and 40 seconds. If the trap
opens too frequently, due perhaps to a cold and windy location, the rate of opening can
be slowed down by the simple expedient of fitting an insulating cover, something which
is standard on high pressure traps. Rapid cycling also occurs when the disk and seat
become worn, and live steam leakage results.

Some manufacturers use a steam jacket to make good the radiation losses from the top
cap. This means that the steam in the control chamber can never condense. A radial
groove is provide on the underside of the disc which allows this steam to leak away.
The problem here is that the groov e constitues a permanent leak. The trap is unable to
give a tight shut-off and tends to work too frequently. The inevitably large steam
jacketed cap is also more wasteful than the provision of an insulating cover on a more
conventional trap.

[Link] Impulse

The impulse trap is shown diagrammatically in Exhibit 6.17. It consists of a hollow

piston A with a piston disc B working inside a tapered piston C which acts as a guide.

At start-up the main valve rests on the seat D, leaving a passage of flow through the
clearance between piston and cylinder and the hole E at the top of the piston.
Increasing flow of air and condensate will act on the piston disc B -and lift the main
valve off its seat to give increased flow. Some condensate will also flow through the
gap between piston and disc, through E and away to the trap outlet.

As the condensate approaches steam temperature some of it flashes to steam as it passes

through the gap. Although this is bled away through hole E it does create an
intermediate pressure over the piston, which effectively positions the main valve to
meet the load.

The trap can be adjusted by moving the position of pisto n C relative to the seat but the
trap is affected by significant back pressure. It has a substantial capacity, in spite of its
small size. However, the trap is unable to give complete shut-off, and will pass live
steam on very light loads. The main problem, however, is the fine clearance between
the piston and [Link] is readily affected by the dirt normally found in a steam
system. Use of this trap is relatively rare, so it is not considered in subsequent
sections.

[Link] Labyrinth

This type scarcely qualifies asan “automatic drain trap” but isshown in Exhibit 6.18.
It consists of a series of baffles which can be adjusted bymeansof a handwheel.

Hot condensate passing between the first baffle and the trap body is subject to a drop in
pressure and some of it flashes to steam. The space around the next baffle has to cope
with an increased volume of hot condensate and flash steam. The result is a restriction

Energy Conservation & Efficiency Project (ECEP)


 




Exhibit 6.17
Impulse Trap

Exhibit 6.18
Labyrinth Trap

 



STEAM TRAPS 6-20

along the length of the trap which slows down the flow of condensate and prevents the
escape of live steam.

The baffle plates can be moved in or out using the handwheel which alters their position
relative to the body. However, the device is fairly crude and requires frequent
adjustment to cope with varying conditions. Again, it is not considered in subsequent
sections.

[Link] Orifice Plate

The theory and construction of the orifice plate as steam trap are extremely simple. By
providing an orifice sized to handle the condensate load, maintenance is simplified and
radiation losses are virtually eliminated. Because of the great difference in density
between condensate and steam, the orifice cannot waste any great amount of steam even
if the sizing calculations are somewhat adrift . There is no question of a valve failing in
the open position.

The main limitation is the size of the orifice. A 1 mm diameter hole can still waste
over 2.3 kg/h of steam under “no load” conditions. It is woefully inadequate to cope
with air at start-up and such a small hole is not easy to protect from dirt.

However, the main danger is waterlogging as the load increases. Although it is claimed
that the fixed orifice can cope with load changes up or down 25%, start-up loads are
commonly 3 or 4 times higher than running loads. Pressure fluctuations are another
problem in calculating orifice size. It is in no sense an automatic device and is not
considered further.

6.3 FACTORS RELATING TO THE SELECTION OF STEAM TRAPS

By definition the steam trap must trap o r hold back steam while at the same time
allowing the removal of air, noncondensibl e gases and condensate.

The selection of steam traps is not always simple, as a trap must be selected to carry out
a given function under given conditions. These may involve variations in operating
pressure, condensate load or back pressure. Traps may be subjected to extremes of
temperature or water hammer, while corrosion or dirt are other common hazards.
Different trap types can have subtly different operating characteristics that are more
important in some cases than others. Finally, steam trap manufacturers are always
making conflicting claims regarding the superior qualities of their traps.

There is no universal steam trap, and trap selection is sometimes difficult. Often,
however, several different traps can be used effectively in a given application. This
section reviews some of the more important factors in trap selection.

Energy Conservation & Efficiency Project (ECEP)


 



STEAM TRAPS 6-21

6.3.1 Air Venting

At start-up the trap must be capable of discharging air. Until air is displaced, steam
cannot enter the steam space and warming up becomes a lengthy process. Standing
losses increase and plant efficiency falls. Separate air vents may be required on larger
or more awkward steam spaces, but in most cases air in a system is discharged through
the steam traps. Here, thermostatic traps have a clear advantage over other types since
they are fully open at start-up.

Float traps with built-in thermostatic air vents are also very good, while the sizable
orifice in some thermodynamic traps means that they too can handle a reasonable
amount of air. The small bleed hole in the inverted bucket trap or the orifice plate
generally means poor air venting capacity.

6.3.2 Condensate Removal

Having vented air, the trap must then handle condensate. If the steam space is small
and output is critical, then condensate must be discharged immediately and at steam
temperature. Waterlogging is one of the main causes of reduced output from a steam-
heated plant. It is also a significant factor in tube plate failure in shell and tube heat
exchangers.

Mechanical traps are clearly the first choice on this count. However, the frequent
cycling of other types may mean that they can be accepted. The discharge of
condensate below steam temperature using thermostatic traps should only be
contemplated when the steam plant being drained can accept a degree of waterlogging.

6.3.3 Thermal Efficiency

If thermal efficiency is defined by how much heat is profitably used in a given mass of
steam, the thermostatic trap may appear to be the best choice. These traps hold back
condensate until it has cooled to somewhat below saturation temperature. Provided that
the heat is given up in the plant itself, to the space being heated or to the process, there
is a real saving in steam consumption . Indeed, there is every inducement to discharge
condensate at the lowest possible temperature.

On the other hand, if cool condensate is then returned to a feed tank which requires
preheating, the “efficient” trap has done little for the overall efficiency of the steam
system.

Care must also be taken in evaluating any application involving a cooling leg . Draining
through a bimetallic steam trap may look attractive in terms of lower temperature
discharge and reduced loss of flash steam. On the other hand, if heat is bein g lost to
atmosphere through an uninsulated cooling leg, then the net gain in thermal efficiency
is probably zero.

 



STEAM TRAPS 6-22

Without a cooling leg, condensate will be held back wi in the heat transfer equipment,
and the main reservation must be whether th e equipment itself will accept this
waterlogging. It is permissible with oversized coils and even desirable with non-critical
tracer lines. As already indicated, however, it can b e disastrous in the case of heat
exchangers.

A way to evaluate the efficiency of operation of the stea m trap i f is related to its
heat loss due to live steam leakage and radiation and convectio n lssses during
operation. This is discussed in detail in Section 6.5.

6.3.4 Water Hammer

Water hammer can damage steam traps and cause them to malfunction. It is therefore
important that this possibility is recognized and the trap selected accordingly.

Water hammer will occur as slugs of water are picked up at high speed in poorly laid
out steam mains or pipe coils. It will also occur when there is a lift after steam traps.
Even when steam pressure appears to be adequate to overcome the lift, problems will
occur at start-up or when the equipment being drained is temperature-controlled.
Discharging condensate into a pumped return line can also produce a type of water
hammer (Section 7.5).

The thin walled bellows or element of the traditional balanced pressure thermostatic
type of trap is clearly vulnerable to damage by water hammer. However, this
possibility is almost eliminated with some stainless steel elements such as the one shown
in Exhibit 6.19. The bimetal trap is nevertheless the more robust trap in those cases
where a thermostatic trap is selected.

Of the mechanical traps, the inverted bucket type is far more resistant to water hammer
than the ball float type. The thermodynamic traps are generally impervious to damage
by water hammer.

6.3.5 D i r t

Dirt is another major factor which must be considered in trap selection. Although
steam condenses to give distilled water, there are many cases where priming in the
boiler causes the carryover of boiler feed treatment compounds. In addition, there are
the problems of initial dirt in the pipe at the time of installation and the products of
corrosion during subsequent running.

A trap having an intermittent blast action is the one least likely to be affected by dirt.
In thermostatic traps this means that the balanced pressure thermostatic trap is to be
preferred, although the larger flat valve associated with some diaphragm traps can cause
difficulties. The dribbling action of bimetallic traps, coupled with the arrangement of
valve stem passing through the seat, means that these are most prone to malfunction or
even to blockage. Claims are sometimes made that a particular element can be readily

Energy Conservation & Eficiency Project (ECEP)


 




Exhibit 6.19
Construction Of Improved
Balanced Pressure Element

 



STEAM TRAPS 6-24

cleanedand is not [Link], fouling of the element is rarely a problem

and thevulnerable part isthe valveand seat.

Mechanical traps are moderately successful. The valve and seat of the float traps are
submerged below the water level which might protect them from dirt at the water
surface or which sinks to the bottom of the [Link] the other hand, the loose sand and
gravel which cause problems when draining concrete-curing autoclaves can be carried
through large float traps quite successfully, no doubt due to the high mass flow.

In the case of inverted bucket traps, the claim is sometimes made that the main valve is
above the water level and unaffected by dirt. This is partially true, but ignores the fact
that the most vulnerable orifice is the air vent hole in the [Link] here can
cause the trap to air bind and fail in the closed Position. It is a well known fact that
some traps operate better after being struck with a heavy object. In the case of the
inverted bucket this can mean dislodging a bit of scale blocking the air vent and
producing an instant cure.

The thermodynamic trap is another type well-suited to dirty conditions, especially if a

strainer is [Link] disc and seat must be clean to ensure correct operation, they
are normally kept in this condition due to the high velocity across them. There is also
the positive snap action. This, incidentally, obviates any chance of wiredrawing as
there is no question of the two parts working closely together. Suggestions that this
type does suffer from wiredrawing are ill-founded. Many thermodynamic traps have
built-in strainers to prevent excessive dirt entering the trap itself.

By contrast, the impulse trap is unsuitable for dirty conditions. The fine clearances
between plug and tapered sleeve do not enjoy the benefits of high velocity flow and the
plug will frequently stick in an intermediate position. In this case, the trap becomes a
fixed orifice, and its improper operation may go unnoticed for a long time.

The orifice plate is of course the device least suited to the dirty conditions. The hole is
almost invariably small and frequently blocks within hours of installation. Enlarging
the hole (as is sometimes done in desperation) makes a nonsense of the original sizing.
It is wasteful and in some cases merely delays the time when blockage will occur. A
strainer is sometimes fitted but this has to be extremely fine if it is to be [Link]
problem then is frequent blockage of the strainer screen.

6.3.6 Steam Locking

The possibility of steam locking should also be considered in the selection of steam
traps. It can occur whenever a steam trap is fitted remote from the equipment being
drained. It can become acute when condensate is removed through a siphon or dip
Pipe.

Exhibit 6.20 shows the removal of condensate from a drying cylinder. In Fig. (i), the
steam pressure is sufficient to lift condensate up the siphon pipe, through the steam trap

 




Exhibit 6 . 2 0

Removal of Condensate form a Drying Cylinder


 



STEAM TRAPS 6-26

and away. Fig. (ii) shows what happens when the level of the condensate at the bottom
of the cylinder falls below the end of the siphon pipe. Steam enters the siphon pipe and
causes the steam trap (in this case a float type) to close.

The trap is thereafter “steam locked”. Heat loss from the cylinder will result in the
formation of more condensate which is unable to reach the trap. Fig. (iii) shows severe
waterlogging of the cylinder which will result in a reduced drying rate from the
cylinder and an increase in the power required to turn the cylinder. In an extreme case,
the cylinder could fill to the center line and cause water damage at the rotary joints.

Exhibit 6.21 shows a float trap fitted with “steam lock release”, This is a needle valve
which allows the steam locked in the siphon pipe to be bled away past the main valve.
It is the only type of trap with this facility and is the only correct choice on such things
as heavily loaded drying [Link] the needle valve is only cracked open to
avoid steam wastage, it has only limi capacity to vent [Link] of this type are
often provided with separate air vents: Exhibit 6.22 shows a trap which combines
steam lock release and thermostatic air vent facilities.

Clearly any trap which opens regularly as a result of heat loss will eventually cope with
a steam [Link], the resultant drainage will be erratic and the method is only
acceptable with small, non-critical condensate lines.

6.3.7 Servicing

The steamtrap incorporates a valve andseat together with movingparts. wear or

fatigue areinevitable, and account must be taken of the need to servicethe trap.

Many traps are designed for easy maintenance. Exhibit 6.23 shows a typical bimetallic
trap where the entire internals can be removed by unscrewing the seat thread. As
discussed earlier, small mechanical traps are sometimes designed so that the cover and
internals can be readily removed to the workshop for attention as shown in Exhibit
6.24.

A recent development is the sealed unit which precludes any kind of maintenance or
proper [Link] the entire trap has to be replaced whenever the internals require
renewal, it is essential that provision be made for this when the trap is installed.

6.4 STEAM TRAP CHARACTERISTICS

Steam trap manufacturers provide a great deal of information on the characteristics of

their specific range of [Link] on the operating characteristics of their models
may differ, but as a general rule most traps of a similar design will perform in a similar
way in service.
Exhibit 6.25 is a summary of the characteristics of the most common types of steam
traps. There may often be situations when tw o or more types of steam trap appear to
meet the system I: [Link] cost of the traps must be taken into consideration,

Energy Conscrvation & Efficiency Project (ECEP)


 




Exhibit 6.21
Float Trap with Steam Lock Release

Exhibit 6.22
Combined Float and Thermostatic
(F&T) Steam Trap

 




Exhibit 6.23
Thermodynamic Trap showing Removal
of Bimetallic Element

 




(b)

Exhibit 6.24
Removal of Steam Trap Internals
(a) Inverted Bucket Trap
Float and Thermostatic Trap

 




HOW VARIOUS TYPES OF STEAM TRAPS MEET SPECIFIC OPERATING REQUIREMENTS

Inverted TD Bellows
Characteristic Bucket F&T (Disc ) Thermoslatic
A Method of Operatiion lntermittent Continuous intermittent (1) lntermittent
B Energy Conservation (Time in Service) Excellent GOOd Poor Fair
C Reststance lo Wear Excellent Good Poor Fair
D Corrosion Rwstance Excellent Good Excellent Good
E Resistance to Hydraulic Shock Excellent Poor Excellent Poor
F Vents Air andl CO2 al Sleam Temperature Yes No No No
G Ability to Vent Air at Very Low Pressure (1/4psig) Poor Excellent No(3) Good
H Ability to Handle Start-up Air Loads Fair Excellent Poor Excellent
I Operation Against Back Pressure Excellent Excellent Poor Excellent
J Resistance to Damage from Freezing (4) Good Poor Good Good
K Ability to Purge System Excellent Fair Excellent Good
L Performance on Very Light Loads Excellent Excellent Poor Excellent
M Aesponsiveness to Slugs of Condensale lmmediate Immediate Delayed Delayed
N Ability to Handle Dirt Excellent Poor Poor Fair
0 Comparative Physical Size Large (5) Large Small Small
P Ability to Handle “Flash Steam” Fair Poor Poor Poor
Q Mechanical Failure (Open - Closed) Open Closed (6) Open (7)
1 Can be contin use on low load 4 Cast Iron traps not recommended 7 Can tail either open or closed
2 Excellent when “secondary Steam isIS utilized 5 In welded stainless sleel construction-medium depending upon the design of the
3 Not recommended for low pressure operations 6 Can tail closed due to dirt bellows

Exhibit 6.25
Comparison of Steam Trap Operating Parameters

 



STEA M TRAPS 6-31

and it is always advisable, in any case, to consult the manufacturer’s representative for
detailed advice.

6.5 ENERGY REQUIREMENTS OF STEAM TRAPS

Since steam traps may sometimes pass steam or lose heat by radiation and convection,
there is interest in the amount of energy that can be lost by different types of steam
trap.

6.5.1 Thermostatic Traps

Under normal conditions, the thermostatic trap holds back condensate until it has cooled
by a certain [Link] does not reach the main valve and so there is no apparent
wastage. Unfortunately, this can lead to waterlogging of the equipment and hence
reduced output. The steam user may increase the consumption of steam to compensate,
but this increase in steam consumption does not appear as an energy requirement
attributable to the steam trap.

The situation can change under no load conditions. Heat loss from the trap body cools
the condensate surrounding the element, causing the valve to open. A small amount of
condensate is discharged and is then replaced by steam. However, due to hysteresis
there is a brief delay before the element responds, and some live steam is lost.
Laboratory tests indicate losses up to about 0.5 kg/h.

Any attempt to insulate a thermostatic trap will result in a significant delay in the
opening of the [Link] can become severe and therefore insulation is not
normally recommended.

6.5.2 Mechanical Traps

A float trap is another example where the valve and seat are normally flooded so there
should be no steam loss through the trap. However, a float trap is relatively large so
there can be a significant heat loss from the trap by radiation. For example, tests on a
12.5 mm trap showed a radiation loss equivalent to 1.4 kg/h of 12 bar steam at an
ambient temperature o f 32.5 C.
o

The inverted bucket trap has been shown to lose about 0.5 kg/h for 12.5 mm traps
under low load conditions. This is due primarily to the steam loss through the vent
hole.

The float trap can and should be insulated, and this will not affect its operation .
Insulating the inverted bucket trap will slow down its operation, which may in some
cases lead to waterlogging. Normally, bucket traps are not insulated.

 



STEAM TRAPS 6-32

6.5.3 Thermodynamic Traps

The thermodynamic trap will lose some steam under low load [Link]
at a temperature close to that of the steam will produce flash steam at the inlet orifice ,
causing the trap to [Link] is on the upstream side and therefore the flooded
valve ensures no loss of steam through the trap. However, heat is then lost from the
cap and the trap will open periodically. Wader low load conditions the condensate on
the upstream side may become exhausted and the trap will then require live steam to
close it.

Much will depend on ambient temperatures but the loss will be around 0.5 kg/h and this
could be doubled in severe weather. On the other hand, losses can be halved by fitting
an insulating cover on the top cap. In any case, these losses disappear as the
condensate load increases, while radiation losses are minimal due to the small size of
the trap. The latter losses are not likely to exceed 0.25 kg/h.

Quantifying the energy requirements of steam traps is not easy. There are many
variables involved, such as ambient conditions, steam temperature and pressure, and the
load on the system. The performance of traps will also depend on size and design.
Exhibit 6.26 summarizes the energy requirements of typical 12.5 mm traps; the figures
are approximate only, and can be used for rough estimates of steam requirements.

6.6 CHECKING STEAM TRAP PERFORMANCE

Because thesteamtrapis a vital link between the steamand condensate systems, correct
operation isessential Three types of failure modes are found in steam traps:

1. Failure in the closed position is usually signaled by poor equipment

performance due to waterlogging with condensate, although this kind of
failure can go unnoticed on steam mains.

2. Failure in the open position causes live steam losses, but is also not always
obvious. Traps will often discharge into a common condensate return
system. Live steam at the collecting tank can indicate problems, but the
defective trap may be difficult to locate.

3. Inefficient operation is the most common failure, but also the most difficult
to identify and locate. It results in steam losses, and can be caused by a
number of reasons specific to trap types. Examples are excessive cycling ,
incomplete closure of the valve, slow closure in response to steam, etc.

The following subsections describe various methods of checking trap performance.


 




NO LOAD REASONABLE LOAD

Thermostatic

Float

1.8.

Thermodynamic

Exhibit 6.26
Energy Requirements of steam Traps

 




STEAM TRAPS 6-34

6.6.1 Traps Discharging to Atmosphere

Traps discharging to atmosphere give the easiest and surest check on their operation.
Under most conditions, even an untrained eye can look at a trap discharge and decide
whether the trap is working correctly . Traps with a blast or intermittent discharge are
particularly easy to check. When the trap is closed only an odd wisp of steam should
be visible, this being caused by the evaporation of any drops of hot condensate left in
the outlet connection. When the trap is discharging, there will normally be an
appreciable quantity of flash steam leaving the outlet with the condensate and this
should not be mistaken for live steam.

Thermodynamic traps, conventional balanced pressure traps and inverted bucket traps
handling moderate loads will all work in this way.

Float traps, bimetallic traps and some balanced pressure traps with stainless steel
elements will in most cases give a continuous discharge. With these types it is more
difficult to tell if they are working correctly . However, if there is a short transparent
zone at the end of the discharge pipe with a bluish hazy appearance, then it is a strong
indication that the trap is blowing live steam.

If a trap discharging to atmosphere is stuck wide open, then live steam loss is apparent
from the noise and accompanying discharge.

6.6.2 Test Vents

One way of overcoming the problems of the closed return system is to fit some kind of
test vent on the downstream side of a trap so that the discharge can be diverted to
atmosphere. The trap is therefor e .discharging open ended and performance can be
assessed along the lines already discussed.

The problem is that the trap is thereby relieved of back pressure. Back pressure will
alter the effective setting of bimetal traps and, if high enough, can render impulse or
thermodynamic traps inoperative , The test is therefore not entirely foolproof and the
results must be considered accordingly.

6.6.3 Sight Glasses

The sight glass shown in Exhibit 6.27 is simply a window fitted on the downstream side
of a trap so that the discharge can be observed. It is generally effective in the case of
traps having a blast discharge since the on/off characteristic is clear. However, some
experience may be needed in judging the discharge from traps having a continuous
discharge, particularly if these are heavily loaded.

Problems can occur on some installations due to deposits on the inside of the glass. It
is as well to remember that deposits may also be affecting the steam side of any heat
transfer surfaces and this is a timely reminder that feed water treatment should be
checked.

Energy Conservation & Efficiency Proje t (ECEP)


 




Exhibit 6.27
Sight Glass

 




STEAM TRAPS 6-36

6.6.3 Temperature

Another old established method of checking steam trams is to measure temueratures at

trap inlet and discharge. Methods ranging from temperature-sensitive crayons to
infrared pyrometers have all been used. Unfortunately these methods are of limited
value. They will detect a trap causing serious waterlogging and may have some
relevance in the case of thermostatic traps. However, the temperature of condensate
and flash steam on the downstream side of a correctly working trap will normally be
around 100°C. This is exactly the same as the temperature of condensate and live
steam on the downstream side of a defective trap. In this situation temperature
measurement gives no guide to trap performance.

6.6.4 Sound

Another well-established method involves listening to the sound made by the trap as it
operates. This no doubt started with a screwdriver handle being used as a make-shift
stethoscope, while the industrial stethoscope is now giving way to a whole range of
ultrasonic leak detectors.

There is no doubt that this method can be effective with some traps. The distinctive
and regular “click” of the thermodynamic trap can be detected by even the untrained
ear. Indeed this clear signal is often an over-riding consideration in selecting the
thermodynamic trap.

Unfortunately there are many traps with no distinctive signal. Condensate and flash
steam immediately downstream ‘of the trap orifice can sound uncommonly like
condensate and live steam at the orifice. Both sounds will be affected by the mass flow
and also pressure. It can be shown that a defective trap under light load conditions
gives a lower signal than a sound trap under full load conditions . A “scan” across the
base of a trap can produce an astonishing spread of different signals . A further
problem is provided by the way in which the sound of adjacent traps can be transmitted
through the pipework. The ultrasonic leak detector can be used, but it may need
careful “tuning” to match or suit trap conditions.

6.6.5 Electronic

A more recent development is provided by a device which uses the electrical

conductivity of condensate. This involves the fitting of a sensor chamber o n the
upstream side of the steam trap. The chamber is shown in Exhibit 6.28 and contains an
inverted weir. With the steam trap working normally, condensate flows unde r this
weir; a small hole at the top equalizes the pressure on each side. A sensor on the
upstream side detects the presence of condensate . By plugging in a portable indicator it
is possible to check whether the electrical circuit is complete and tha t the trap is
working properly.

If the trap fails in the open position then a relatively large volume of steam wil l flow
towards the trap. This depresses the water level on the upstream side of the weir and

Energy Conservation & Efficiency Project (ECEP)


 




Trap working correctly

(Green light showing on indicator)

Live Steam

Trap passing Steam

(Red light showing on indicator)

Red and Green Lights

Exhibit 6.28
Operation of Spira-Tec (Spirax)
Detector for Steam Traps

 




STEAM TRAPS 6-38

exposes the sensor. The electrical circuit is broken and the portable indicator signals
that the trap has failed.

The advantage of this system is the very definitive signal which can be interpreted
without resorting to experience or personal judgment. Using suitable wiring, the test
point can be fitted remote from the sensor chamber. This can be useful in the case of
traps at high level or in ducts which would otherwise be inaccessible . A disadvantage
is the additional cost of this sensing system for each trap.

6.6.6 Ultrasonic Leak Detectors

Ultrasonic leak detectors and steam trap testers convert high frequency sounds emitted
by steam traps into frequencies that are audible to human hearing. The ultrasonic
device is able to eliminate background noises and “tune-in” on the steam trap alone . As
such, the ultrasonic steam trap tester is a “contact” device: the sound transducer must
be placed in contact with the device being tested.

Different types of steam traps make different types of characteristic noises while in
operation. An operator or technician can be trained to recognize these sounds and
diagnose a failed steam trap. For instance, typical conditions encountered in common
steam traps are as follows:

1. Inverted bucket traps normally fail in the open position. This results in a
continuous rushing noise as the steam blows through the steam trap. The
bucket may sometimes also be heard banging against the side of the steam
trap body.

2. Float and thermostatic traps normally fail in the closed position. A pinhole
leak in the ball float will cause the float to be weighted down, or water
hammer will collapse the ball float. Since the trap is failed closed, no
sound will be heard. Alternatively, if the trap has failed in the open
position, the continuous rushing sound of steam blowing through the trap
will be heard.

3. Thermodynamic traps generally fail in th e op en position. a l l o w i n g

v

continuous blow-through of steam . If the trap normally, the

ultrasonic detector can pick up the clicking of the disk as it cycles 4 to 10
times per minute.

4. Thermostatic traps failed closed will be silent, while those failed open will
result in the continuous rushing sound of steam. In normal operation, the
ultrasonic detector will be able to pick up the sound of the valve cycling
open and closed.

Energy Conservation & Efficiency Project (ECEP)


 



STEAM TRAPS 6-39

6.6.7 Other Methods

It is worth mentioning one indirect method of trap checking. Although steam metering
is not as common as it should be, there are those plants where the steam consumption is
metered. In those cases where the load is steady it is possible to arrive at a standard
figure when everything is operating correctly.

A significant rise in steam consumption is an indication that the traps are beginning to
pass steam. Since steam metering will usually apply to a relatively large machine or
area, it leaves the problem of deciding which particular traps are giving trouble.
However, it is a useful prompt that some more individual checking is required.
Alternatively, it may indicate the need for wholesale repair or renewal.

6.7 TROUBLESHOOTING

The summary given in Exhibit 6.29 provides general guidance in locating and
correcting steam trap problems. Many problems are in fact steam system problems
rather than trap problems.

6.8 STEAM TRAP MAINTENANCE

Maintenance is rarely a prime consideration when selecting steam traps and is

frequently neglected. The textbook approach suggests steam traps should be examined
and, if necessary, serviced on a regular basis. It is an approach used on most large
items of plant equipment but is rarely extended to steam traps.

In many cases, repair of a trap is not cost effective or indeed cannot be done due to the
particular trap design. However, balanced pressure thermostatic traps often have an
element which is designed for easy replacement and therefore changing these on a one
to two year schedule may well be justified (depending on the trap service and
performance [Link] are available for the replacement of parts in many inverted
bucket [Link] on thermodynamic traps can also be easily replaced, although if not
properly matched with the original trap, may give poor results.

A regular check of performance of all traps in a plant is obligatory. Depending on

plant operation and trap size, the checking should range from a weekly to a bimonthly
basis. Every plant should maintain a complete log of all steam traps, with dates
checked and maintenance recorded. If trap performance is suspect, the trap should be
replaced; the cost is normally paid back by the steam saved in as little as one year.
Replacement is generally cost-effective for all but the largest traps; in those, the repair
of the internals may be practical.

 




Cold Trap-No Discharge.

Steam Loss. If the trap blows live Steam. C. Trap may have insufficient air-
if trap fails to discharge condensate.
trouble may be due to any of the
then following causes
handling capacity or the air may
not be reaching trap in either case
A use auxiliary air vents.
A Valve may fail lo Seal
1. Piece of scale lodged in Orifice
1 Wrong pressure originally Mysterious Trouble. If trap Operates
2. Worn parts satisfactorily when discharging to
specified.
2. Pressure raised without in-
stalling smaller orifice
I B I B Trap may lose its prime
1. If the trap is blowing Iive s t e a m
atmosphere bul trouble is encounleree
when connected with return line. Check
3. PR V out of order close the inlet valve for a few the lollowing
4 Pressure gauge in boiler reads minutes Then gradually open
A Back pressure may reduce capacity
low if the trap catches its prime the
5. Orifice enlarged by normal wear of trap
chances are that the trap is all
6. High vacuum in return line 1 Return line 100 small - trap hot
right 2 Other traps may be blowing
increases pressure differential 2. Prime loss is usually due to
beyond which trap may operate steam - trap hol
sudden or frequent drops in 3 Atmospheric vent in condensate
sleam pressure. On such jobs receiver may be plugged - trap
B No condensate or steam coming
the installation of a check valve hol or cold
to trap.
is called f o r - location D or C 4 Obstruction in return line - trap
1. Slopped by plugged strainer in Fig 39.1 possible locale
ahead of trap. hot
trap well below drip point 5 Excess vacuum in return line
2 Broken valve in line to trap
3. Pipe line or elbows plugged. - trap cold
C. For F & T and thermostatic traps
thermostatic elements fail lo close lmaginary Troubles. If it appears that
C. Worn or detective mechanism.
Repair or replace as required. steam escapes every lime trap dis.
Continuous Flow. If an I B or disc trap charges. remember Hot condensale
D. Trap b o d y filled with dirt discharges continuously or an F & T or forms flash sleam when released to
Install strainer or remove dirt al thermostatic trap discharge al lull lower pressure. but it usually
capacity. check the following. condenses quickly in the return
line
E. For 1.B bucket vent filled with dirt. A. Trap too small
Prevent by: 1. A larger trap. or additional traps
1. Installing strainer should be installed in parallel
2 Enlarging vent slightly 2. High pressure traps may have
3. Using bucket vent scrubbing been used for a low pressure rob
wire. Install right size of internal
mechanism
F. For F & T Traps if air vent is not
functioning properly trap will likely B. Abnormal water conditions
air bind. Boiler may foam or prime
throwing large quantities of
G. For Thermostatic Traps. the bellows waler into steam lines A
element may rupture from hydraulic separalor should be installed
shock causing the trap to fail closed or else the feed water con-
ditions remedied
H. For DISC Traps trap may be installed
backward. Sluggish Healing. When trap operates
satisfactorily, but unit fails to heal
Hot Trap-No Discharge. properly:
A. No condensale coming lo trap.
1. Trap installed above leaky by- A. One or more units may be short.
pass valve. circuiting and the remedy is to
2. Broken or damaged syphon pipe install a trap on each unit See
in syphon drained cylinder. page 14
3. Vacuum in water heater coils may
prevent drainage install a B. Traps may be too small for rob Source:
vacuum breaker between the even though they may appear to Steam Condensate, Guidelines
heal exchanger and the trap. be handling the condensale ef- for Condensate Drainage (M101)
ficiently Try next sized larger A r m s t r o n g M a c h i n e Works, Three
trap. Rivers, Michigan, USA

Exhibit 6.29
Troubleshooting Steam Traps

 




7. CONDENSATE RECOVERY

7.1 ENERGY SAVINGS

Steam condensate is usually high purity water at a moderate to high temperature. In all
plants, the maximum amount of condensate possible should be returned for reuse i n the
boiler. Where the boiler feedwater temperature is low in a plant, the reason shoul d bc
found. It could be due to a low rate of condensate return, lack of insulation o n the
condensate return pipes or to losses from the feed tank, either as heat through th e wall
or as overflow.

As much condensate as is economically nossible should be returned from source s where

there is no likelihood of contamination. This will save heat, make-up water and any
chemicals used in water treatment and will reduc e blowdown losses. The saving of fuel
resulting from increased condensate return may be estimated from Exhibit 7.1.

If, for example, the feed water temperature can be raised by increasing the amount of
o
condensate returned at 90 C from 25 to 100 per cent. a saving of about (12. 5 - 3.5 =
9) percent of the boiler fuel will result (Exhibit 7.1) . Further savings will accrue due
to the accompanying reduction i n blowdown losses.

Where there is a danger of feed water contamination, automatic dumping by measuring

the conductivity of the condensate is common. The monitors should be sited so that
only the contaminated condensate is dumped and not the whole of the condensate. In
large installations it may be iustified to monitor and dump each condensate source
independently. Without-special arrangements it is rarely possible to use feed water in
excess of 82’C due to cavitation problems on the feed water pump, though the feed
water temperature can be raised above this level after the pump where an economizer is
fitted. When in doubt the boiler feed pump manufacturer should always be consulted.

Collecting condensate means taking the discharge from one or more steam traps into a
piping system. It is important that the piping does not interfere with th e proper
operation of the steam traps. The designer of a condensate return system should
therefore consider the various elements described in subsequent sections:

1. general layout

2. pipe sizing

3. lifting condensate

4. flooded return lines

5. condensate handling.

Energy Conservation& Efficiency Project (ECEP)


 




Exhibit 7.1
Boiler Fuel Saved by Condensate Return

 




CONDENSATE RECOVERY 7-3

-

7.2 GENERAL LAYOUT OF CONDENSATE RETURN SYSTEMS

In a well-designed system, condensate return will impose no unreasonable back pressure

on the steam traps. This means the provision of adequate pipework, arranged so that
condensate can flow away by gravity. In the ideal layout the fall will be sufficient to
overcome the frictional resistance of the pipe when carrying the maximum flow. Traps
relieved of back pressure will be able to discharge condensate even at the low pressures
associated with start-up.

A proper condensate return system is also important for good air venting capability. It
is of limited use having a trap with good air venting capacity if the air then has to
bubble its way through a length of flooded pipe. The partially flooded pipe of the
gravity system ensures that there is free space through which the air can escape.

It is rarely possible to run condensate by gravity all the way back to the boiler feed
tank. It is therefore more usual to run condensate to collecting points from which it can
be pumped back to the boilerhouse. Such an arrangement has two clear advantages:

1. The traps retain the benefits of no back pressure already outlined above.
This is particularly important when what could be lengthy run s back to a
central boiler plant are considered.

2. The collecting tank or receiver incorporating an atmospheric vent is a useful

indication of the condition of the traps discharging into the tank. The
release of live steam from traps requiring attention may seem wasteful but it
may be less harmful than pressurizing the whol e condensate return system
and affecting the drainage from other items of plant . The vent identifies the
area where trapping problems are occurring so that they can be investigated
and corrected.

A persistent discharge plume from the condensate receiver may also confir m the
presence of worthwhile amounts of flash steam and point the way towards some sort of
flash recovery scheme (Chapter 8).

7.3 PIPE SIZING

The accurate sizing of condensat e pipework is extremely difficult. Under normal

working conditions hot condensate will be reduced in pressure as it passes through the
trap orifice. Some of the condensate will flash to steam and the condensat e pipework
will be expected to carry this flash steam along with the condensate.

Although the percentage of flash steam is relatively small in terms of mass, it occupies
a large volume. On the other hand it would be reasonable to expec t flash steam to
travel at a higher velocity than the condensate . Clearly, any calculation of this complex
two-phase flow would require consideration o f different pressure and temperature
conditions. A simple compromise will cover most situations.

Energy Conservation & Efficiency Pmject (ECEP)


 



CONDENSATE RECOVERY 7-4

At start-up the plant is cold steam will condense rapidly and the condensate pipework
will be required to pass two or three times the normal condensate rate However, there
will be little or no flash steam due to the relatively cool state of the condensate. As the
plant warms up, the amount of condensate reduces to the running load. As the
condensate temperature goes up, however, so too does the amount of flash steam
formed at the trap orifice.

Experience has shown that if a pipe is sized as if it were carrying water under the
starting conditions, it will have adequate capacity to handle the condensate and flash
steam under the running conditions. For this kind of exercise the start up load should
be taken as not less than twice the running load. This fairly simple rule of thumb will
cover applications involving pressures up to 7-10 bar .

For higher pressures it is worth increasing pipe sizes in order to handle the larger
volumes of flash steam. This is especially the case with traps draining steam mains
which will tend to operate at maximum pressures and where start-up loads can exceed
estimates.

The carrying capacities of pipes in sizes up to 100 mm are detailed in Exhibit 7.2. The
following example shows how the table can be used.

Example

If a plant is known to have a running load of 750 kg/h then for the purpose of condensate pipe
sizing we can say that we must design for a starting load af 1500 kg/h.

Total back pressure and length of travel are also important. Referring to Exhibit 7.2, 1500 kg/h
of water through a 25 mm pipe has a frictional resistance of about 3.5 mbar (350 Pa) per meter
of travel.

If the distance from trap discharge to condensate receiver or boiler feed tank is no more than
about 3 m, then the back pressure can be considered insignificant. If, on the other hand, the
condensate has to travel over a length o f 500 m, then the back pressure will be 50 0 x 3.5 =
1750 mbar (175 kPa) which, allowing for the additional resistance of pipe fittings, will impose a
total back pressure of about 2 ba r (200 kPa), to which must be added the back pressure caused
by any lift in the line.

Whether the back pressure in the example is acceptable depends upon the steam
pressure available at the trap. It is obviously not advisable to impose too high a back
pressure especially as this is normally at its highest during start up when the steam
pressure is at its lowest.

High back pressure cannot be tolerated at all where the equipment temperature is
regulated by a modulating type of control. On the other hand, condensate lines should
not be oversized otherwise cost of installation can be prohibitive. A larger capacity
steam trap may have to be sized in order to provide adequat e flow at the higher back
pressure.

 




Paper inbar 15mm 20mm 25mm 32mm 40mm 50mm 65mm

metre per m kg/h

28 0.28 90 209 380 865 1320 2254 5194 8079 16511

30 0.30 93 218 403 890 1361 263 5348 8314 17000
0.33 97 226 414 930 1420 2744 5579 8688 17736
33
39 0.39 107 249 469 1028 1565 3025 6142 9526 19514
43 0.43 113 265 496 1079 1446 3180 6454 10024 20457
45 0.45 116 270 508 1107 1687 326 6618 10297 21002
47 0.47 119 277 521 1134 1729 3338 6777 10523 21500
0.50 123 286 538 1172 1787 3447 6949 10859 22154
50
53 0.53 127 296 557 1211 1346 3565 72.35 11249 22907
0.55 139 302 569 1238 1887 3638 7380 11476 23380
55 11703 23814
57 0.57 133 308 580 1261 1923 3710 7525
135 314 591 1288 1959 3783 7666 11884 24258
59 0.59 24121
61 0.61 138 320 502 1311 1996 3851 7806 l2lll
0.64 141 327 615 1338 2041 3933 7970 12383 2,5220
64 25991
67 0.67 146 337 534 1379 2400 405 1 8210 12746
148 343 645 1402 2136 4119 8342 12973 26400
69 0.69 26863
0.70 149 345 649 1411 2150 4146 8432 13041
70
71 0.71 150 348 655 1424 2168 4182 8473 13154 26868
73 0.73 152 354 665 1447 2200 4246 8600 13336 27216
0.75 154 358 673 1456 2227 4219 a695 13517 27506
75 29732
78 0.78 158 366 689 1497 2282 4390 8900 13812
80 0.80 160 370 595 1510 2300 4427 8972 13925 28350
82 0.82 162 375 704 4529 2331 449 9072 14407 28758
88 0.88 168 391 733 I590 2427 4536 9453 14651 29885
90 0.90 170 395 740 1606 2449 4717 9548 14787 30142
98 0.98 179 414 777 1696 2567 4944 10025 15513 31616
l00 1.00 180 418 785 1701 2590 4990 101I5 15649 31879
1.14 194 440 845 1832 2790 5306 10841 16329 34247
114 34746
118 1.18 198 457 857 1860 2830 5443 11022 17055
1.20 199 462 867 1 880 2860 5502 111I3 17282 35720
120 36742
131 1.31 209 484 907 1996 2994 5761 11657 18053
1.37 215 497 931 2018 307 5906 11948 18507 37667
137 38012
1.40 216 502 939 2037 3103 5965 12066 18688
140 19187 39055
147 1.47 224 516 966 2096 3189 6128 12383
157 1.57 231 534 1002 1168 3298 6337 12814 19822 40361
160 1.60 234 541 1011 2195 3334 6409 12973 20049 40797
1.63 237 546 1025 2219 3370 5477 13109 20276 41232
163
176 1.76 246 570 I066 2309 3511 6740 13608 21092 42938
1.80 249 576 1075 2331 3547 6808 13744 21319 43364
180 45360
196 1.96 261 603 1129 2440 3710 7130 14379 22317
265 611 1143 2472 37S0 722 1 14560 22589 45931
200 2.00 47265
212 2.12 273 629 1179 2549 3874 7434 15014 23270
2.16 275 634 1189 2167 3905 7493 14132 23451 47637
216
220 2.20 278 641 1200 2595 3942 7570 15277 23678 48101
2.29 284 655 1225 2649 4028 7729 15604 24177 49125
229
2.35 288 664 1243 2689 4086 7843 15840 24522 49832
235 50363
240 2.40 292 672 1256 2719 4129 7927 16017 24780
295 679 1270 2749 4173 8010 16193 25039 50994
245 2.45 51929
255 2.55 301 694 1297 2906 4260 8176 16520 25556
701 1311 2834 4304 8260 16683 25814 52445
260 2.60 304
305 703 1315 2944 4318 8287 16738 25900 52619
261 2.61
2.74 313 721 1348 2917 4331 9501 17173 26554 53039
274 26717 54432
277 2.77 315 727 1356 2935 4459 9555 17292
730 1363 2948 4479 8593 17357 26839 54536
280 2.80 317
325 749 1402 3025 4595 8813 17781 27533 55793
294 2.94 56426
3.00 328 757 1414 3055 4641 8900 17956 27803
300 57607
310 3.10 335 771 1438 3112 4726 9063 18280 28305
1446 3129 4742 9074 18390 29459 57879
314 3.14 336 775 58424
3.20 340 784 1462 3163 4825 9204 18579 28767
320 29076 53963
327 3.27 344 792 1479 3198 4853 9299 18779
348 901 1495 3230 4904 9408 18979 29366 39512
333 3.33 60057
351 809 1511 3263 4955 9516 19178 29656
340 3.4

E x h i b i t 7.2

ater FI ta

 



CONDENSATE RECOVERY 7-6

In many situations it is reasonable to size on a frictional head loss in the region of 0.8
mbar (80 Pa) per meter of travel, subject to a reasonable back pressure especially where
low pressure or very long lines are concerned.

7.4 LIFTING CONDENSATE

Although condensate should preferably run from a trap by gravity, there are inevitable
situations where condensate has to be “lifted”. The question is often asked whether a
particular trap will lift condensate. The true answer is that no steam trap will “lift
condensate”. - The condensate is lifted by the pressure available at the trap outlet. For
each 0.11 bar (11) kPa) pressure at the trap, the condensate can be lifted approximately
1 meter.

It must be remembered that the pressure at the trap outlet is often much lower than the
steam supply pressure. Temperature control, whether automatic or by hand, means
restricting the steam supply so that the lower steam temperature gives the reauired
reduced heat output. This can mean very low steam pressures, and zero press&e or
vacuum conditions are not uncommon in some heat exchangers. Even where there is
no attempt at temperature control, the rapid condensation of steam at start-up can cause
very low pressures at the trap outlet.

If condensate has to be lifted, wate r hammer with its attendant noise and risk of
mechanical damage is inevitable. Steam traps should be selected with this in mind.
The robust inverted bucket trap is a good choice for heat exchangers where float traps
might otherwise be preferred.

The back pressure will of course reduce the differential across a trap and thereby reduce
its capacity. This should be borne in mind when reading the sizing charts.

7.5 FLOODED RETURN LINES

There are few problems connecting the discharge from any number of traps into a
common condensate collecting system providing that th e pipework is properly sized.
Problems do occur, however, when condensate is taken into a flooded return line.

This commonly occurs when draining steam mains. It is quite usual to find pumped
condensate lines following the same route and there is a temptation to put the discharge
from mains drain traps into such lines.

As the high pressure condensate from the steam mai n passes through the trap, part of it
will re-evaporate into flash steam . The expanding flash steam must find a place in the
return line- which is already full of water-at a lower pressure. It can only enter by
forcing water out of the way. However, these steam bubbles will collapse suddenly in
the relatively cool condensate causing severe hammer.

The best solution is to avoid the flooded line and return the condensate independently to
the nearest collecting point as shown Exhibit 7.3 . Where this is not practical it is

Energy Conservation & Efficiency Project (ECEP)


 




.
Stepm

Condensate

Exhibit 7.3

Steam Traps Should Not Discharge Into

Flooded Condensate Return Lines

 




CONDENSATE RECOVERY 7-8

advisable to use a trap which will hold back the condensate until it has cooled. This
can lead to waterlogging of the steam mains, so an adequate drain pocket and a 2-3 m
unlagged cooling leg is essential. Alternatively, a float trap with continuous discharge
may be acceptable. This will not allow the condensate to cool but the steady discharge
can usually be absorbed into the flooded line without serious problems.

7.6 CONDENSATE HANDLING

Where condensate has to be discharged over a long distance it is advisable to reduce the
back pressure due to frictional resistance as far as possible. In Exhibit 7.4 this is done
by arranging an initial vertical lift followed by a gravity run back to the boiler feed
tank. This is preferable to the alternative arrangement of long horizontal run, followed
by a final vertical lift which ensures that th e pipework is completely full at all times.

Where condensate is collected in a receiver it is frequently necessary to pump it to a

higher point, such as the boiler feed tank. Both centrifugal and positive displacement
pumps can be used.

Problems with centrifugal pumps can occur when they are called upon to handle hot
condensate. Condensate at high temperature can flash to steam, drastically reducing the
capacity of the pump, while the collapse of steam bubbles when in contact with the
impeller can cause severe vibration and erosion. Pumps and piping systems must be
designed to cope with conditions of low net positive suction head.

Some of the problems encountered by centrifugal pumps are solved by the use o f a
simple positive displacement pump, such as the “Spirax Ogden Automatic Pump”. Its
method of operation is described in Exhibit 7.5, while Exhibit 7.6 shows a typical
installation.

The vented receiver ensures a constant hea d during the time that the pump body is
filling. During the discharge stroke it acts as a reservoir so that the flow of condensate
from the plant is not interrupted. The operating medium is usually steam. Compressed
air can be used but this may aerate the condensate and aggravate corrosion.

Properly installed, this type of pum p can give long periods of trouble-fre e service.
There is no cavitation and it will pump condensate at boiling point . Because there is no
motor it is unaffected by moisture: it will continue to operate when completely
submerged, a common enough hazard in some pits used to collect condensate.

Since the pump handles a measured amount at each stroke, it is relatively easy to
calculate the flow rate by counting the number of strokes . By adding a simple stroke
counter to the top cover, the pump can be used as a simple metering devic e to measure
the condensate load of the connected area or plant. The flow counte r is shown Exhibit
7.7.

It must be remembered that most condensate pumping systems are in operation

intermittently, so the discharge piping must be sized accordingly. In the case of the
Ogden pump, the cycle is divided between tilling time and emptying time. Filling time

Energy Conservation & Efficiency Project (ECEP)


 




Exhibit 7.4

Proper Method of Condensate Discharge

to Overhead Tanks

 




Operation
1 Liquid flows by gravity into pump through inlet
check valve A, starting to fill body. Float B moves
upwards along exhaust valve spindIe C liquid
cannot pass through nutlet check valve D because
back pressure holds valve closed Exhaust valve E is
kept open (and Inlet valve F closed) by weight
of C. Float B continues to rise until it presses
against callar G, raising spindle and so partially
opening inlet valve F and partially clasing exhaust
valve E.
2 Pressurlsed operarting medium flowinq in through
inlet valve F causes exhaust valve E to snap closed
and F to open fully Pressure to tranrmitted through
liqiuid to inlet valve A. closing It. Pressure !n pump
then builds up until just sufficient to overcome
back pressure.
3 Pressurised liquid opens outlet check valve D and
discharge commences.
4 Float B falls with liquid level until. wlrh most of
Iiquidd pumped out. It engages with bottom collar H
of exhaust valve spindle C. Weight Of float pulls
exhaust valve E open, to vent the pump body. and
simultaneously, Inlet valve F closes Without
pressure in pump body. outlet check valve closes
and Inlet check valve will open to gravity flow of
Iiquid.

Exhibit 7.5

 




Condensate from Vent

Steam Trap draining steam supply

Exhibit 7.6
Typical Installation of Steam Powered
Positive Displacement Condensate Pump

Exhibit 7.7
Condensate Flow Counter for
Positive Displacement Pump

 



CONDENSATE RECOVERY 7-12

is generally twice as long as emptying time so the discharge pipework should be sized
to handle two to three times the rated capacity .

An electric pump set is rather different in that receiver size and pump capacity should
be selected to give pumps which run fairly continuously under full load conditions and
which avoid too frequent starting and stopping at other times. In many such cases the
discharge pipework should be sized to handle 1.5 times the return rate. As with the
Ogden pump, the rate must be taken into account when calculating the frictional
resistance of the discharge pipework .

7.7 CORROSION

Corrosion can quickly damage a condensate system if careful attention is not

maintained. Inevitably, over a long period of time, every condensate system will suffer
the attacks of corrosion.

Corrosion is caused primarily by carbon dioxide entering the system through the
feedwater. It may also be formed in a zeolite type of water softener from sodium
bicarbonate. When steam is generated, the CO 2 is distributed along with the steam
through the piping. As the steam condenses, the CO 2 stays in equilibrium with the
condensate. But as the condensate .begins to cool, its ability to absorb CO2 increases,
and the carbon dioxide enters into solution, generating a carbonic acid (H2 O + CO 2 =
H2CO3).

The CO 2 concentration increases with the cooling of the condensate, and the carbonic
acid can reach a pH of 4; its corrosivity increases by the higher temperature. The acid
attacks copper and iron pipe, and the presence of oxygen further increases its corrosive
effect. A list of points which will help minimize condensate system corrosion is as
follows:

o Maximize condensate return

o Maintain returned condensate at the highest temperature possible

Avoid use of thermostatic traps which significantl y subcool the condensate

Use a deaerator

o Maintain adequate venting of heat exchangers

o Use proper chemical feedwater treatment .

Energy Conservation & Efficiency Project (ECEP)


 



8. FLASH STEAM

8.1 INTRODUCTION

The previous chapter considered the collection of condensate and its return to the boiler
feed [Link], it did not take into account the important problem of flash steam
which is discussed in detail below.

The problem arises from the relatively high temperature of [Link] we consider
o
a steam coil using steam at 7 bar then steam will condense to water at 170. 5 [Link] the
coil is fitted with a float trap, condensate will be released at steam temperature.
Thermostatic traps will hold back condensate until further cooling has taken place but
the discharge temperature will still be well above 100 C.
o

When the pressure of this condensate falls, in the pipework or at a collecting tank, the
condensate can only exist at the saturation temperature corresponding to the pressure in
the [Link] excess temperature provides the heat which re-evaporates part of the
condensate to form flash steam. This steam in turn condenses in the atmosphere to
form the familiar clouds of steam wherever hot condensate is discharged.

The careful designer will take account of this problem and ensure that flash steam is
collected for re-use. Flash steam is as useful as steam from the boiler; using it will
reduce steam demand. It will prevent the loss of untreated water and eliminate the
clouds of white vapor which are often synonymous with steam waste.

8.2 ESTIMATING THE QUANTITY OF FLASH STEAM

To make use offlash steam we nee d know howmuch is available. This can be
determined fromfirst principles or can read from asimple table.

Consider the jacketed vessel shown in Exhibit 8.1. With the aid of steam tables
(Appendix A) the quantity of flash steam can be calculated as shown in the following
example:

The condensate arriving at the steam trap under a pressure of 7 bar gauge will have an enthalpy
of saturated water of 697.22 kj/kg At 0 bar gauge (IOOoC) the maximum enthalpy of saturated
water is 417.46 kj/kg; the difference is 279.79 kj/kg.

This difference of 279.7 9 kJ/kg is the surplus enthalpy which is converted to enthalpy of
evaporation in evaporating some of the condensate.

At atmospheric pressure, steam has an enthalpy of evaporation of 225 8 kJ/kg With 279.79
kJ/kg available, the amount of flash steam formed at atmospheric pressure from each kilogram of
condensate is:

( 2 7 9 . 7 9 kj/kg/ I2258 kj/kg) x 1 kg = 0.124 kg

Energy Conservation & Efficiency Project (ECEP)


 




__--
e-.-

--.-------
----

Condensate 7 bar gauge 170.5oC

100 psi 3380F

Exhibit 8.1
Flash Steam is Generated From
High Pressure Condensate

 



FLASH STEAM 8-3

If the plant isusing 1000 kg/h of steam, the amount of flashsteamformedatatmospheric

pressure each hour is 1000 x 0.124 = 124 kg.

It must be realized that if the steam trap draining the unit is of a type that does not
discharge condensate as soon as it forms, some of the sensible heat in the condensate
will be given up before it passes to the lower pressure and so the amount of flash steam
will be reduced. This is shown in the following example.

Example

If in the example above, the trap was of the thermostatic type, discharging condensate to
atmosphere at 15-C below the steam temperature (164.97 - 15 = 149.97). the enthalpy of
saturated water of the condensate on discharge would be approximately 632.2 kJ/kg,
corresponding to a saturation temperature of 149.97.C. The surplus heat available for producing
flash steam would be reduced to:

632 - 417.46 = 214.54 kj/kg

From 1000 kg of condensate per hour, the flash steam production would be:

(1000 kg x 214.54 kj/kg) / 2258 kj/kg = 95 kg/h of flash steam

This example shows the way in which the amount of steam can be calculated,
depending on the temperature of the [Link] steam will be formed whenever
condensate at a given pressure is subjected to a reduction in pressure. Exhibit 8.2 gives
a direct reading of how much flash steam is formed whenever this pressure reduction
takes [Link] assumes that the traps are discharging condensate at steam temperature.
If condensate is discharged below steam temperature, then it is better to calculatethe
amount of flash steam using the method described in the last example.

8.3 FLASH VESSEL DESIGN .

Flash steam is formed at the point where pressure drop takes place,generally
downstream of the steam trap orifice. From this point onwards the condensate return
system must be capable of carrying this flash steam as well as condensate. As
described previously, many condensate return systems suffer high back pressure
because of inadequate allowance for flash steam in pipe sizing.

Once the flash steam is formed, the next operation is to separate it from the condensat e
using a flash [Link] is usually a vertical vessel with the condensate and flash
steam entry point part on the side of the vessel. The vessel is sized such that a
considerable drop in velocity takes place when the condensate/flash steam mixture
enters. As a result the condensate falls to the bottom of the vessel where it is drained
[Link] height above the inlet allows flash steam to be taken off at the top outlet
without picking up any droplets of water which may rise into the steam space.

Prompt drainage is essential, so the ball float steam trap is almost invariably the firs t
choice for installation at the condensate outlet .

Energy Conservation & Efficiency Project (ECEP)


 




Quantity of Flash Steam

15
14

13

12
11

10
9

0.14 0.18

kg Flash perkg Cmdsnutr

Exhibit 8.2
Quality of Flash Steam Generated From
Condensate Pressure Reduction

 




FLASH STEAM 8-5

Flash vessels covering a wide range of condensate recovery applications are available
commercially. Typical details from a manufacturer are shown in Exhibi t 8.4.

Alternatively, a flash Vessel can easily be constructed locally . Using Exhibit 8.3 , the
following example shows how such a vessel can be designed.
Example

A plant operating on steam at 11.0 bar gauge condenses 4000 kg/h o f steam. Flash from this is
to be recovered at 0.5 bar gauge. The amount of flash can be calculated from steam tables or by
reference to Exhibit 8.2, which shows the condensate at 1 1 .O bar gauge produces just under 0.15
kg of flash steam per kg of condensate when the pressure is droppe d 0 0.5 bar gauge. From
4000 kg/h of condensate, 600 kg/h of flash steam will be formed.

From Exhibit 8.3, 4000 kg/h of condensate meets the line of 600 kg/ h of flash steam in the sire
300 area, so a size 300 flash vessel, as identified in Exhibit 8.4, is ne cessary for this duty.

The vessels shown in Exhibit 8.4 can cope with loads of up to 9,000 kg/h. Vessels for
large duties can be designed by observing some basic points:

1. A reasonable flash steam velocity must be maintained in the vessel itself.

To aid separation and to minimize the carryover of water droplets the
velocity of flash steam passing up through the vessel should not exceed 3
m/s. With a knowledge of the amount of flash steam to be handled and the
volume of this steam at the designed pressure, the vessel diameter can be
calculated.

2. It is equally essential to cope with the condensate. There will be

conditions when a relatively larg e condensate flow will produce a small amount
of flash steam. Working on steam velocity alone might give a small diameter
vessel with highly turbulent water rushing out the bottom. Exhibit 8.5 shows
minimum vessel diameters based on condensate load. To select a vessel the
larger of the two diameters should b e used Both figures should be regarded as
approximate guides and there is no harm in using a larger diameter if such a
vessel or tube is more readily available.

3. Flash vessel recommended heights and sizes of the condensate outle t ar e

also given in Exhibit 8.5. The flash steam outlet should be sized on a steam
velocity not exceeding 15 m/s. The condensate inle t will usually be fixed
by the proper sizing of the condensate line . The job of the flash vessel is to
separate the flash steam and condensate, and it helps if high velocity and
turbulence are avoided by having a generously sized line . As a general

Energy Conservation & Efficiency Project (ECEP)


 




2.5 - 9 000
380

1.25- !4500
-----300
Example
I

0.63 _ 2 250

0.25 - 900

o - 0
225 450 900 1 400
Flash Steam kg/h
I
0 0.060.13 0.25 0.4
Flash Steam kg/s

Exhibit 8.3
Sizing of Condensate Flash Vessels

 




Exhibit 8.4
Typical Flash Vessel Specifications

 



FLASH STEAM 8-8

rule, the condensate inlet should be at least one pipe size larger than the
flash steam outlet.

4. A typical flash vessel is shown in Exhibit 8.6. Note that the inlet
connection is one third of the way up the vessel. This allows space so that
any surge of condensate does not obstruct the inlet while the distance above
any water surface will minimize the carryover of water droplets.

Exhibit 8.5

Sizing of Flash Vessels -- Diameter, Height and Condensate Outlet Size

Maximum
Condensate Diameter of Recommended Condensate
Quantity Flash Vessel Height Outlet Size
(kj/hr) (mm) (mm) (mm)
9,000 380 1100 35
12,500 460 1200 50
16;000 500 1400 65
20,000 600 1400 65
34,000 760 1400 80
50,000 920 1500 100

8.4 DESIGN OF FLASH STEAM RECOVERY SYSTEMS

The successful application of flash steam recovery depends on three factors being met :

1. The first essential is clearly a supply of condensate at a reasonable pressure.

At the same time the traps supplying the condensate must be able to accept
the back pressure which will be created by the operating pressure of the
flash steam system.

2. The second requirement is a suitable user for the low pressure flash steam.

3. The proposed use for flash steam should be close to the source.
Transporting low pressure steam will involve relatively large pipework.
Radiation losses and the installation costs can be high, and may outweigh
the advantages of flash steam recovery .

Provided that these conditions are met, the next design consideration is the method of
control. Given a supply and demand which are in step, and a demand which exceeds
the available supply, the simplest system is to pipe the flash steam to the intended user
without the need to provide additional steam.

Exhibit 8.7 shows this kind of arrangement, where flash steam is used to preheat air in
a multi-bank heater [Link] very adequate heat transfer surface of the flash steam
battery and its position across he air inlet will ensure the rapid condensation of any

 




Flash Outlet

Connection

Condensate Outlet

Exhibit 8.6
Typical Flash Vessel Design

 




Exhibit 8.7
Utilization of Flash Steam
Flash Steam Condenses at a Pressure Related to the Load

 




FLASH STEAM 8-11

flash steam. The flash steam pressure will find its own level but the pressure is
unlikely to be more than atmospheric.

This arrangement has the advantage of simplicity, and the hig h pressure traps are not
subjected to any back pressure at start up. On the negative side, the heat transfer
surface of the flash steam section will not be fully utilized at time of low flash steam.
In addition, there can be problems in draining the flash vessel due to the very low
differential across the flash vessel steam trap.

Exhibit 8.8 shows the same arrangement but with a make-up reducing valve. This
supplies any deficiency on the flash steam side and effectively “controls” the flash
steam pressure. It ensures that there is always a useful contribution from the flash
steam battery when there is little or no flash steam and that there is a reasonable
differential across the trap on the flash vessel. The only problem is that the high
pressure traps will have to start up against a small back pressure, although this ca n be
avoided by shutting off the make up supply until the main batteries are up to pressure.

8.5 SOME TYPICAL APPLICATIONS

Systems where flash steam supply and demand are in step are clearly the ideal.
Exhibits 8.7 and 8.8 are systems of this type. It is when the heater batteries are called
upon to supply heat that flash steam becomes available . It can then be condensed in the
first battery which is essentially an air pre-heater.

Similar arrangements can be made for large areas heated by unit heaters or radiant
panels. Ten to fifteen percent of the heaters are separated and supplied with flash steam
generated from condensate collected from the remaining heaters. since all or none of
the area has to be heated, supply and demand are again in step.

This means that the output of the unit heaters or panels will be reduced, owing t o the
lower steam temperatures available in the flash steam section. However, this can
normally be accepted. If not, then a small increase in the heating surface will be
necessary to compensate.

Exhibit 8.9 shows a system where condensate from the process plant is used to provide
flash steam to augment the heating supply. This may be perfectly satisfactory for much
of the year but flash steam will be blown to waste at any time when heating is not
required (for example, in the summer) . This is a case where supply and demand are
not in step. The arrangement is not ideal, but the savings to be mad e during the heating
season may well justify the cost of the installation.

An interesting and useful self-contained method of flash recovery is to be found in what

is known as a packaged hot water generator unit (Exhibit 8.10). The to p is a normal
steam to water hot water generator for space heating or process hot water. Condensate
and flash steam discharged from the steam trap are passed to the lower unit where they
are separated, the flash being condensed in the upper section of the unit and the
condensate being sub-cooled in the lower section. These two heat exchangers are
placed in series in the secondary circulation so that return water picks up usefu l heat

Energy Conservation & Efficiency Project (ECEP)


 




Exhibit 8.8
Utilization of Flash Steam
Make-up Steam Controls Pressure in the Flash Steam System

 




Condensate

Exhibit 8.9
Process Flash Steam used in Coils
for Plant Heating

Ogden Pump

Exhibit 8.10
Process Flash Steam
Used for Hot Water Heating

 



FLASH STEAM 8-14

from the flash steam and condensate, thus reducing the demand for live steam. This
again is an application where flash steam supply and demand are in step but the flash
vessel is replaced by a modified shell and tube preheater.

 



9. HEAT LOSS FROM STEAM SYSTEMS

The reduction of heat loss from steam and condensate systems through the use of the
proper insulation is a practical means of achieving substantial energy savings in many
plants. Insulation can often be applied easily and quickly without interrupting an
operation or process.

To reduce undesirable heat losses, many different types of insulation are commercially
available, with different properties and operating temperature ranges. Costs can vary
widely. In selecting insulation, a number of factors should be considered:

1. thermal properties at the operating temperature

2. mechanical properties and susceptibility to mechanical damage

3. fire resistance

4. resistance to water, chemicals, oil, and other materials

5. cost to purchase and cost to install.

Comprehensive data are usually available from manufacturers. Typical data on

commonly-used materials are given in Exhibit 9.1.

Exhibit 9.1

Examples of Insulating Materials

Thermal
Conductivity
Material Form Max. Temp.
o
C W/mOK

Glass fiber Fiber slabs 510 0.048(l)

85% magnesia Preformed slabs 310 0.058(l)
Calcium Silicate Slabs 650-1010 0.061(l)
Calcium Silicate Slabs 650-1010 0.085(2)
Mineral fibre (rock) Loose fill 760-950 0.054(I)
Silica fibres Loose fill 980 0.063(2)
Isocyanurate Foam 145 0.024(3)
Polyurethane Rigid foam 100 0.024(3)
o o o
(l) at 200 C (2) at 600 C (3) at 20 C

The effect of insulation in reducing heat losses may be estimated using various
equations for heat transfer by conduction, convection and radiation. However, this
manual will not cover the theoretical aspects of heat loss but will focus on simplified
methods based on graphs and tables.

Energy Conservation & Efficiency Project (ECEP)


 




HEAT LOSSES FROM STEAM SYSTEMS 9-2

9.2 HEAT EMISSION FROM UNINSULATED PIPES

Simple tables of heat loss from bare horizontal pipes are presented in Exhibit 9.2.
These data may be used to estimate losses from steam and condensate piping in a
factory wherever insulation has not been applied. As a rough guide, insulation can be
economically justified to reduce these losses by at least 90%.

Using Exhibit 9.2 and knowing the per mete r cost of pipe insulation, an approximate
but rapid check of the economic viability of insulating a bare pipe can easily be made.
o
This chart is based on ambient air at 20 C and no wind. Where windy conditions
prevail, Exhibit 9.3 provides some factors for estimating the effect of wind speed on
heat loss.

The cost of insulation can be compared with the value of 90% of the heat being lost
from the bare pipe. Provided a reasonable payback period is indicated, a more detailed
evaluation of the proposed insulation will be justified. For this, the graphs described in
the following sections of this manual should prove useful. Where only a short length of
piping is being investigated and the payback is clearly very rapid, detailed investigation
will probably not be necessary.

Exhibit 9.2

Heat Loss from Bare Pipes

Heat Emission from Bare Horizonta l Pipes, Ambient Temperatures 10-C - 21 oC, Still Air Conditions

Temp. Diff .....................................................pipe Sim (mm) ........................................................................

Steam to Air 15 20 25 32 40 50 65 80 100 150
oC
Watts per linear meter

56 54 65 79 103 108 132 155 188 233 324

67 68 82 100 122 136 168 198 236 296 410
78 83 100 122 149 166 203 241 298 360 500
89 99 120 146 179 205 246 289 346 434 601
100 116 140 169 208 234 285 337 400 501 696
111 134 164 198 241 271 334 392 469 598 816
125 159 191 233 285 321 394 464 555 698 969
139 184 224 272 333 373 458 540 622 815 1133
153 210 255 312 382 429 528 623 747 939 1305
167 241 292 357 437 489 602 713 838 1093 1492
180 274 329 408 494 556 676 808 959 1190 1660
194 309 372 461 566 634 758 909 I080 1303 1852

Finally, it is important to note that it is not only th e pipework that may need insulation.
It is common practice to finish the insulation abou t 80-100 mm (3 to 4 inches) from
pipe flanges or valve flanges, leaving valves and flanges uncovered. This is generally
done for a number of reasons. For example, a lack of insulation makes access to the
flanges easier. Also, without insulation, leaks can be easily detected, and leakage
should not corrode the flange bolts. However, uninsulated flanges are a sourc e of
significant heat loss. Valves have a large surface area and lose proportionally more
heat; these too should be insulated.

Energy Conservation & Efficiency Project (ECEP)


 



HEAT LOSSES FROM STEAM SYSTEMS 9-3

Exhibit 9.3

Effect of Wind Speed on Heat Lass

Wind Speed (km/hr) Relative Heat Loss

Still Air 1.0

1.0 1.5
10 2.0
14 2.5
19 3.0
26 3.5
34 4.0

Easily removable clip-on insulation (shaped for flanges and valves) is available, or it
can be constructed.A bad leak will soon make itself known, regardless of insulation.
In fact, insulation is one of the better ways to prevent leaks, since bare flanges on hot
insulated pipes introduce temperature stresses at the flanges which may cause leaks.

For estimating heat losses from valves, an uninsulated valve can be regarded as losing
the same amount of heat as 1 meter of bare pipe of the same diameter. An uninsulated
flange loses the equivalent of 0.5 meter of bare pipe of the same diameter .

Example

A 100 meter length of uninsulated 150 mm diameter steam pipe has a surface temperature of
150’ C. The ambient temperature is 20’ C. The system operates 24 hours per day. It is
proposed to insulate this pipe with 50 mm thick preformed rigid fibrous sections for an estimated
cost of LE 5,000. Fuel for the 75% efficient mazout boiler costs LE 50 per tonne with a heating
value of 43 kJ/tonne What is the payback period for the insulation?

Calculation

1) Heat loss per meter of bare pipe (interpolating in Exhibit 9.2) is 1028 W/m

2) With 50 mm this drops to 103 W/m reduction

loss)

3) Savings are thus (1028 - 103) = 92.5 W/m or 92,500 Watts for the full length of
Pipe

4) This is equivalent to: (92,500 W x 3600 Joules/W/h x 24 h/day)

= 7.99 x 109 Joules per day = 7.99 GJ/day

5) In terms of boiler fuel, 7.99/0.75 = 10.65 GJ/day

6) The equivalent cost of mazout is (LE 5O/tonne) / (43 kJ/tonne) = LE l.l6/GJ

7) Cost of heat lost = 1 0 . 6 5 GJ/day x LE l.l6/GJ = LE 12.35/day.

payback period: LE 5,000 / LE 12.35/day = 404 days


 



HEAT LOSSES FROM STEAM SYSTEMS 9-4

9.3 HEAT LOSS FROM INSULATED PIPES

9.3.1 Insulating Materials

A wide range of insulation is available from many different companies. The heat loss
graphs in this section are based on four common product types:

1. preformed rigid fibrous sections, including rock and glass fibers

o
2. preformed rigid calcium silicate or, for up to 300 C, 85% magnesia
sections

3. preformed rigid polyisocyanurate and phenolic or, for up t o lOO*C,

polyurethane sections

4. expanded nitrile rubber and polyethylene foam sections.

All new insulation products are asbestos-free. Asbestos should not be used as a
component of insulation because of the serious health hazard posed b y asbestos
[Link]’s literature should always be consulted to determine the properties
and performance of their products.

9.3.2 Economic Thickness of Insulation

Individual plant circumstances must be taken into account to calculate the amount of
insulation that can be justified economically . Each plant will have different fuel and
insulation costs, and different boiler efficiencies. Criteria for the payback period o r
rate of return, against which proposed capital investments must be judged, will be set
by management.

One method of evaluating insulation projects, based on payback periods , is as follows:

1. Determine the cost, in LE, of insulation for different insulation thickness.

2. Determine the heat loss for each level of insulation from the appropriat e
manufacturers literature.

3. Calculate the corresponding energy savings, taking into account the boiler
efficiency and the cost of fuel.

4. Calculate the payback for the lowest level of insulation and compare it with
the factory criterion.

5. If the payback is shorter than the factory requires, calculate the incremental
cost to insulate to the next level and the corresponding incremental value of
savings. Complete the incremental payback and compare with th e factory
requirement.

Energy Conservation & Efficiency Project (ECEP)


 



HEAT LOSSES FROM STEAM SYSTEMS 9-5

6. If the incremental payback is still better than the requirement, repeat the
procedure for the next increment of insulation, until the incremental
payback fails to meet the factory requirement. The last increment which
meets the requirement is acceptable and this represents the economic
thickness of insulation.

This calculation can be carried out using the internal rate of return (IRR) as the
criterion set by management: instead of payback, the initial IRR and incremental IRR’s
should be calculated using a discounted cash flow technique, based on an agreed project
life.

Other methods of establishing the economic thickness of insulation are given in the
following reports:

o Economic Thickness of Industrial Insulation, Conservation Paper 46, 1976.

Available from US Government Printin g Office, Washington, DC 20402
(Stock number 041-018-00115-8).

o * The Economic Thickness of Insulation for Hot Pipes, Fuel Efficiency

Booklet No. 8, UK Department of Energy, Thames House South, London
SWIP 4QJ, England.

Energy Conservation & Efficiency Project (ECEP)


 



10. CHECKING STEAM SYSTEM OPERATION

10.1 INTRODUCTION

The “energy performance” of a steam system may be evaluated by testing the system
for:
1. steam pressure and temperature

2. improperly functioning steam traps

3. condensate recovery

4. leaks and missing or insulation.

Taking the boiler and steam distribution system together, it is often useful to ask the
following questions:

Steam Generation

1. Do we have the right boilers?

2. Are the boilers working efficiently and correctly?

3. Are the fuel and steam metered? for each boiler?

4. Are the feed tank arrangements satisfactory?

5. Is blowdown under control?

Steam Distribution

1. Is the system layout as good as it should be?

2. Is the system effective?

3. Is steam trapping satisfactory on steam mains?

Steam Utilization

1. Is the equipment used properly?

2. Are temperatures properly controlled?

3. Is drainage and air venting correct?

4. Do the steam traps waste steam?

5. Are steam traps working?

Energy Conservation & Efficiency Project (ECEP)


 



CHECKING STEAM SYSTEM OPERATION 10-2

Condensate Handling

1. Is condensate properly used?

2. Are there other losses or problems?

3. What about flash steam?

Comments on each item are offered below; and for further details, the appropriate
sections of this manual should be consulted.

10.2 STEAM GENERATION

A detailed review of this topic is outside the scope of this manua l .In brief, however,
the questions to be asked are:

1,. Do we have the night boilers?

* Boilers should be properly sized for the steam load on the syste m lightly
loaded boilers will have significant radiation losses. Burners should also b e
properly sized and able to respond efficiently to different load conditions.

2. Are the boilers working efficiently and correctly?

The major losses for any boiler are represented by the hot combustion gases
discharged at the stack. Gas analysis and a check on stack temperature will
indicate the level of combustion efficiency being achieved in a boiler. The
operating pressure in the boiler steam drum should be near the design value :
low pressure operation can promote water carryover into the steam distribution
system.

3. Are the fuel and steam meterd? for each boiler?

Most plants have data on the monthly or annual cost of fuel but it is far from
common to find fuel metered to individual boilers. The fuel consumption for
each boiler is important information for effective energy management.

Metering of steam production is, unfortunately, not carried out by many plants.
Sub-metering of steam flows in different sections of the steam distribution
system is all too [Link] lack of meters makes the calculation of a steam
balance extremely difficult (if not impossible) and does not permit the proper
management of steam consumption by different users/departments in a factory .
Without steam meters, proper allocation of costs to the different users cannot be
made, nor can improvements achieved by energy conservation be evaluated. In
many plants, metering of steam using simple orifice meters gives adequate
means of monitoring consumptions, provided proper attention is paid to the
correct installation of the meter, and care is taken to adjust the meter factor for
actual operating conditions of temperature and pressure.

 




CHECKING STEAM SYSTEM OPERATION 10-3

4. Are the feed tank arranments satisfactory?

This refers to the boiler feed tank, which provides a reservoir for the returned
condensate and fresh water make-up. The operation of the system around the
tank should be checked, and the temperature of the “mixed” water monitored.
Is the feed tank receiving all of the available condensate? What is the
percentage of cold make-up? If condensate is being pumped back at a nominal
o
100 C, is the actual feed temperature what should be expected, given the
addition of cold feed?

If condensate is returned directly, is it properly sparged into the tank? Vapor

loss is often the result of letting condensate fall through space into the feed tank
rather than being injected below the water level, On the other hand,
unreasonably high temperatures may indicate that steam traps are passing live
steam.

Is the feed tank properly sized? To cover fluctuations and possible interruptions
in supply it is normal to hold enough water to provide one hours steam at
maximum rating. But is there enough free space or ullage to cope with the
relatively large condensate return at start-up?

In some cases the temperature of the feed water is limited by the ability of the
feed pumps to handle it. The problem is cavitation, and can often be avoided by
increasing the head over the pump.

Because the feed tank is hot, steps must be taken to minimize heat losses. The
greatest losses will take place from the water surface . Some form of cover or
lid is therefore essential. Insulating the tank will provide additional savings.

5. Is blowdown under control?

Blowdown must be carefully controlled to an optimum level. Too little

blowdown may cause solids buildup in the boiler, and result in scaling; too
much blowdown will waste energy. Dissolved solids content of the boiler water
should be measured regularly, and the blowdown adjusted accordingly.
Maximizing condensate return, which has no impurities, helps to minimize
boiler blowdown.

10.3 STEAM DISTRIBUTION

Major questions to be asked are as follows:

1. IS the system layout as good as it should be?

Does the system need to be modified or piped in a different way? Even in the
best-operated plant the steam distribution system represents an energy loss. Can
this loss be reduced?

Energy Conservation & Efficiency Project (ECEP)


 




CHECKING STEAM SYSTEM OPERATION 10-4

Incorrect pipe sizing is one obvious source of waste. Steam distribution is no t

an end in itself and the need to supply good quality steam at the right pressure
must be paramount. Nevertheless, a 75 mm line will involve 50% greater heat
loss than a 50 mm line, other factors being equal.

The problem is that replacement of lines can be expensive. Due to the

possibility that loads might be increased at some future date, a degree of
oversizing can usually be tolerated. But is all pipe sizing acceptable? In
extreme cases pipework is left open to live steam even though it no longer
serves any useful purpose. Such pipework should be isolated or removed.

Should the entire system be replaced to meet current needs? After all, boilers
and process plant may be replaced from time to time.

Are heating and process requirements properly separated? Is steam the best way
to provide heating ? Although a central heat source has much to recommend it,
significant distribution losses may be avoided by the simple expedient of
supplying a small local heater, perhaps operated by electricity, to a single
remote area of the plant.

2. Is the system effective?

The sole purpose of the steam distribution system is to supply steam when
required in the right condition and in the right quantity.

Is start-up satisfactory? Inadequate air venting is a common cause of slow start-

up. Steam traps with good air venting characteristics may be fitted, but their
capacity may be limited if the air has then to bubble through long lengths of
flooded condensate line. Thermostatic air vents at the terminal ends of mains
and branches will ensure prompt air venting.

Condensate must be removed promptly. Does the system accord with good
practice in having a fall towards adequate drain pockets? Are branches taken
from the tops of mains? Are the water pockets formed by concentric reducers,
large globe valves or wrongly-fitted strainers avoided?

Inadequate drainage can mean partially flooded lines. This makes a nonsense of
correct sizing and the result is increased pressure drop and steam starvation at
the user end. Even more serious is the-risk of water hammer: fast-moving
steam can pick up slugs of accumulated water . The resultant water hammer is a
prime cause of steam leaks. In other cases accumulated water can damage valve
packings, again leading to wasteful leaks. (For a further discussion of steam
leaks, see Section 10.6).

The formation of condensate is reduced by insulating the steam distribution

pipework. But is insulation adequate ? High fuel costs mean that high standards
of insulation can be justified.

Energy Conservation & Efficiency Project (ECEP)


 




CHECKING; STEAM SYSTEM OPERATION 10-5

Special attention should be given to uninsulated flanges and valves. Lines run
out of doors or in places where insulation can be affected by water should be
carefully checked. Most insulation depends for its effectiveness on minute air
cells. It can become useless if these are filled with water. Waterproofing is
therefore essential.

Finally, remember that many applications benefit from the supply of the best
quality (dry) steam. Would the use of a separator improve plant performance?

3. Is steam trapping satisfactory?

The provision of adequate drain pockets and an otherwise satisfactory steam

distribution system can become worthless if the steam trapping itself is not
correct.

Are the traps of the correct type and properly sized? Traps should be capable of
removing condensate quickly under varying load conditions, and especially
under the low pressure conditions which occur at start-up. First principles
suggest mechanical traps with the inverted bucket trap best able to cope with
water hammer conditions. However, the small, robust thermodynamic trap,
which opens at frequent intervals and loses a minimum of heat from the trap
body, is often preferred, especially for outdoor installations.

Sizing may seem an unnecessary consideration since 15 mm traps have sufficient

capacity for all but the largest lines . However, is this true in the area of the
boiler house? Here traps may be coping with an unknown amount of carryover
from the boilers. Larger traps may well be required to ensure a proper steam
supply to the system.

Once correctly chosen, are the traps working ? Surveys often show a number of
traps out of use. In most cases, condensate is simply carried on to the next
drain point. However, such arrangements are not acceptable and more than one
water hamm er disaster has been traced back to this kind of malpractice.

The reason for these problems may be dirt, or simply neglect. Downstream
traps have to cope with scale and dirt from aging pipework, while those near the
boiler become the resting place for boiler feed treatment compounds carried
over into the steam mains. Blocked traps, strainers or even pocket s filled with
solids arc commonly found.

The blowing trap is an obvious concern for the energy manager, Is proper
provision made for checking stea m traps? The common thermodynamic trap
gives a fairly regular and audible “click” as it closes. Its performance can be
checked by even an untrained ear. Other systems for checking traps are
available, including the provision of a sight glass on the downstream side,
although under some conditions of load some experience is needed to understand
the flow nattern observed. In some cases. a vent to atmosnhere on a three-way
valve system downstream of the trap is sometimes provided. This is used to
take the trap discharge to the open where it can be observed. The problem is

 




CHECKING STEAM SYSTEM OPERATION 10-6

that the trap is then effectively relieved of any back pressure, so the operating
conditions are not entirely normal. Again, experience is extremely useful in
interpreting the observations.

10.4 STEAM UTILIZATION

Most aspects of steam utilization are beyond the scope of this report. In brief, the main
questions relating to steam system efficiency are:

1. Is the plant used properly?

Is steam shut off to equipment which is not being used? When equipment is in
use, is it properly loaded? Drying equipment which is only partially filled will
waste energy. In some plants, the mechanical removal of water (between
rollers, or in centrifuges, for example) prior to further drying by thermal means
may not be as complete as it should be . A small improvement in the moisture
removal by mechanical means can result in a significant reduction in steam
demand.

2. Are temperatures properl y controlled?

Two basic cmestions should be asked . First, have the correct temperatures been
selected? Second, are these being correctly maintained? Control systems that
have gone out of adjustment, or that have been tampered with for production
reasons, are a prime cause of energy waste.

3. Are drainage and air venting correct?

A similar question has already been discussed in the context of steam

distribution. It is equally relevant to steam utilization (refer to Section 10.3,
item 2).

4. Do the steam traps waste steam?

This question is generally asked about thermodynamic traps. The

thermodynamic trap closes as condensate approaches steam temperature. It
therefore closes with water on the upstream side and under these conditions it
cannot waste steam.

Problems can only occur when the regular opening of the trap has cleared this
accumulated condensate. If there is a low load condition and no more
condensate is present, then steam will be required to operate the trap.
Consumption or steam wastage will depend on the actual conditions but could be
up to 1.5 kg/h. Furthermore, as the trap gets worn, the seating of the valve is
not as close, and it may begin to leak some steam; worn traps also cycle much
faster, increasing the losses of steam during opening and closing.

Energy Conservation & Efficiency Project (ECEP)


 




CHECKING STEAM SYSTEM OPERATION 10-7

The truth is that under no load conditions every type of trap will “waste” steam,
and every worn trap will be less efficient than a new one. Thermostatic traps
hold back condensate and apparently avoid steam loss. They only open when
the water in the trap body is cooled by radiation -- a form of steam loss. The
problem is that unless the water on the upstream side is replenished, steam .will
reach the trap. Since the trap must eventually open, steam will be “wasted”.

Steam trapping is a compromise. Losses through traps only occur under low
load conditions, while radiation losses take place all the time. Actual amounts
are small, and the fact that replacing old traps by new ones (of any type) will
reduce steam consumption can confuse the issue. Replacement simply
eliminates the loss through defects and wear, which are far more significant than
differences between trap types.

5. Are steam traps working?

Trap performance is a major concern for achieving good energy efficiency.

This topic is discussed further in Section 6 of this manual.

10.5 CONDENSATE HANDLING

This topic is of particular relevance to those responsible for operating steam distribution
systems.

1. Is condensate properly used?

If steam is used at 7 bar then the condensate contains approximately 25% of heat
used to produce the steam. If the condensate is wasted then the efficiency of
steam utilization must be something less than 75 % .

Condensate is hot distilled water which is ideal for use as. boiler feed water.
Returning it to the boiler feed tank will save considerable fuel. It will also save
water, the cost of water treatment, and heat wasted in excessive blowdown.
Condensate is commonly “returned” to the boiler feed tank, but does it get
there?

Some feed tanks allow condensate to overflow to the drain at start-up. Is this
happening at collecting points ? It is all very well to provide condensate return
pumps but if these have insufficient capacity to cope with start-up loads then
condensate can be lost to drain.

Is cavitation taking place? It is not uncommon to find pumps “running” but

with condensate over-flowing to drain. The pumps are incapable of handling
hot condensate which flashes to steam at the pump inlet. The result is a rather
noisy pump which rotates but handles little or no condensate. Should pumps
with special impellers be used, or is there a case for substituting a different type
of pump which will handle boiling condensate?

Energy Conservation & Efficiency Project (ECEP)


 




CHECKING STEAM SYSTEM OPERATION 10-8

One reason given for not returning condensate is the risk of contamination.
Perforated coils in an acid vat or oil tank might allow these harmful substances
to get into the boilers where the damage would be considerable. In many cases,
such risks have been obviated by improved design. However,. where
contamination is a real possibility, can reasonable precautions be taken?
Mechanical or electrostatic filters can be used to cope with oily condensate.
Detectors can be used to signal the presence of harmful acids or salts. The risk
of returning contaminated condensate should rarely be an excuse for dumping
condensate to drain.

If discharge to drain is inevitable, can some of the useful heat be extracted? It

has long been recognized that recovering heat from hot effluent from textile dye-
houses or laundries makes economic sense. Heat is often recovered from the
condensate from flash vessels used to recoup come of the losses from boiler
blowdown.

2. Are there other losses or nroblems?

Before looking at problems it is worth asking whc:ther all available condensate is

being returned. Condensate collection is often a question of picking up
seemingly insignificant quantities.

In other cases, condensate is effectively lost because steam is used by direct

injection. Is this the best way of using steam? Producing hot water by injecting
steam into a tank is cheap and easy and makes full use of the heat in the steam.
But would it make more sense to use a coil so that the condensate could be
returned to the boiler? This would reduce water treatment costs and might also
reduce the energy costs arising from boiler blowdown.

What of the collecting system itself? This will usually consist of piping from
trap outlets to the feed tank or to condensate receivers . Is it properly laid out,
properly insulated and is it actually being used?

3 What about flash steam?

Condensate systems are designed to handle condensate and varying amounts of

flash steam. Condensate may be collected at the feed tank or at the receivers of
condensate return pumps, but what happens to flash steam? The clouds of vapor
or flash steam associated with many steam systems are unsightly and can
damage the fabric of buildings. More to the point, they represent the loss of
heat as well as treated water. Have these losses been quantified?

The solution to some flash steam problems may be simple . In the case of the
feed tank, the answer may be to take the return below the water surface so that
any flash steam will be condensed. A vacuum breaker will be necessary to
prevent water being drawn back at shut-down . There must also be enough cold
make-up water to condense the flash steam.

Energy Conservation & Efficiency Project (ECEP)


 




CHECKING STEAM SYSTEM OPERATION 10-9

In most cases it is preferable to recover flash steam and utilize it close to the
source. This practice is well established. Condensate from high pressure steam
users is taken to a separating vessel which supplies flash steam to low pressure
users. Can low pressure users be found or created?

In the case of a steam heating system for example, it may be reasonable to cut
off the steam supply to 10-15% of the heaters and supply these with flash steam.
The output from these units will be reduced but the margin on existing output
will usually ensure adequate heating . In the case of larger heater batteries it is
sometimes possible to use flash steam in some kind of preheater. This may
mean installation of a new section of the battery or the adaptation of an existing
one.

Flash steam condensers are a reminder of the fact that supply and demand
should be in step. In some flash recovery schemes this is not the case, and there
are times when flash steam is blown to waste. This might seem wasteful but it
is better to save some flash steam some of the time rather than letting it all go to
waste all the time.

Can better utilization be devised, and is it genuine flash steam which is being
wasted, or is it live steam blowing past failed steam traps? An advantage of the
flash recovery system is that a defective trap will simply blow steam and
augment the flash steam supply. In moderation, this can provide reasonable
insurance against excessive waste . If it leads to the neglect of trap maintenance
then it can lead to imbalance and wastage.

Again, is the flash system working as intended? Is a make-up reducing valve

required? If one is fitted, is it set correctly? Too high a pressure may mean
unacceptable back pressure on traps.

10.6 STEAM LEAKS

There are various ways of estimating the quantity of steam lost through steam leaks.
Exhibit 10.1 is a typical graph which will allow rough estimates of losses to be made,
and thus losses can be quantified in money terms. Where losses are large, it may often
be justified to shut down parts of the system for immediate repairs, rather than waiting
for a scheduled maintenance shutdown some months in the future.

Energy Conservation & Etficieney Preject (ECEP)


 




2 4 6 8 10 14
Steam pressure bar (x100=kpa)

Exhibit 10.1
Steam Loss Through Leaks (Orifices)
Vented to The Atmosphere

 




11. INSTRUMENTATION FOR STEAM MEASUREMENT

In this chapter, we address the subject of steam measurement and monitoring. The
chapter includes a discussion of common pressure measurement instruments, and flow
measurement instrumentation.

11.1 PRESSURE

Pressure -- defined as force per unit area - - is a parameter whose measurement is

important from a process control and energy monitoring point of view. Reasons to
measure pressure include: (1) process control, where product quality depends upo n
certain steam pressures being maintained for specific lengths of time ; (2) safety, where
boiler or reactor pressures should not excee d certain maximums; and (3) energy
measurement, where, for example, the steam pressure indicates its energy content . In
this section, we discuss three basic types of pressure measurement instruments most
suitable for steam: bourdon tubes, diaphragm and capsular elements, and
piezocapacitive strain gage elements.

11.1.1 Bourdon Tube Pressure Gauges

Bourdon tube gauges are widely used and are available for pressures ranging from 1 to
700 bar. These gauges are characterized by low cost, simple construction, good
accuracy/cost ratio, and adaptability to electrical transducers. The main disadvantages
are loss of precision at pressures below 3.5 bar and a susceptibility to damage from
shock and vibration.

These gauges are based on the simple mechanical principle that a closed end, coiled
tube will tend to straighten out when a gas or liquid under pressure is allowed to enter
the tube. Movement of the coiled tube, called a Bourdon tube, is translated by geared
sectors and pinions into the proportional movement of a pen or pointer on the pressure
measuring instrument for recording or indicating. The Bourdon movement can also be
coupled electronically to various transmitters and transducers.

There are three types of Rourdon tube gauges, each of which is described below and
illustrated in Exhibit 11.1.

"C" Bourdon Tube Pressure Gauge: The C shaped Bourdon tube is mostly
used in round case pressure gauges that ar e connected directly to process vessels
and lines for local pressure indication. The tube is flattened by an amount
dependent upon the material of the tube and the required pressure range. When
pressure is applied to the open end, th e tube straight ens out; this movement of
the tube is converted to pointer travel.

Spiral Bourdon Tube Pressure Gauge: The spiral is a thin walled tube
flattened on opposite sides to produce an elliptical cross section. Pressure
applied to the open end uncoils the tube. The several turns of the spiral provide
a large movement of the free end which makes it ideal for use in a recorder.

Energy Conservation & Efficiency Project (ECEP)


 




Bourdon tube’

Pressure

“C* Bourdon

Spiral Bourdon

Helical pressure tube

Exhibit 11.1
Three Types of Bourdon Tube Gauges

 




INSTRUMENTATION FOR STEAM MEASUREMENT 11-3

C: The pressure sensing element consists

of a tube which has been flattened on opposite sides and then formed into a
helix. Pressure applied to the inside of the tube causes the helix to uncoil. This
motion actuates the pen or pointer linkage.

Bourdon tube gauges are normally installed at the measurement location and are directly
connected to the process by a tube. These gauges can be installed in both horizontal
and vertical positions; however, they are sensitive to movement and should be installed
at locations where there is little or no vibration.

11.1.2 Diaphragm and Capsular Element Pressure Gauges

Diaphragm and capsule type pressure elements are used for intermediate an d low
pressure ranges of 0 - 14 bar. These gauges are good in accuracy, moderate in cost,
small in size, and linear in response. The capsules are constructed from a wide variety
of materials for service at temperatures o f -240oC to 540 oC. These gauges can be
installed on a control panel or at the point of measurement.

The diaphragm and capsular pressure sensing elements are illustrated in Exhibit 11.2.
The diaphragm is a flexible disk, usually with concentric corrugations a s shown and is
made to exacting specifications. The diaphragm can be used independently as a
pressure sensor, but is also the basic component of a capsular element. A capsule
consists of two diaphragms, joined together around their peripheries by soldering or
fusion welding, and then stress relieved . Capsules are of two types: convex, in which
the orientation of the corrugations of the two diaphragms is opposed; and nested, in
which the orientation of the corrugations matches.

Pressure deflects the diaphragm or capsule, which in. turn moves a shaft. A multiple
capsule arrangement, having a number of similar elements, can be used when greater
output motion is desired. A flexible plate or similar device restricts the travel of the
element; therefore, a capsular element typically can withstand an over pressure of at
least twice its full range value. These gauges are directly connected t o the process by a
pipe or a tube.

11.1.3 Strain Gauge Pressure Transducers

‘In this type of pressure transducer, pressure changes are translated into changes in
electrical signal. The measuring range is 0 to 700 bar. These gauge s are quite
accurate, have quick response making them suitable for high frequency dynamic
pressure measurements, have a long life and long term stability. They are not sensitive
to orientation, but are limited to low temperatures (generally belo w 300 oC) and
noncorrosive environments. Strain gauge pressure transducers are manufactured in
many forms for measuring gauge, absolute, and differential pressures and vacuum.
Their cost is moderately high.

The strain gauge consists of a wire grid bonded to the surface of a small slip o f rag
bond or Bakelite impregnated paper. When the paper is subjected to load, the wire is

Energy Conservation & Efficiency Project (ECFP)


 




Convex capsul.

Corrugated Flexib
corrugated Flexible
surface
Surface

__--- free-end fitting ---

_.’
spacer ‘.

_
Capsules

Exhibit 11.2
Diaphragm and Capsular Element
Pressure Gauges

 




INSTRUMENTATlON FOR STEAM MEASUREMENT 11-5

stretched, changing its electrical resistance . Measurement of the change in resistance

results in the evaluation of the pressure. A strain gauge pressure sensor is shown in
Exhibit 11.3.

Precise measurement of the strain gauge’s resistance is accomplished b y means of a DC

or AC bridge. The bridge receives its power from the transformer incorporated in a
potentiometer. This is used to determine the extent of imbalance of a bridge when the
strain gauge resistance changes from its normal value.

Strain gauge pressure transducers are screwed into a threaded hole in the location at
which a pressure reading is required . The transducer does not allow local indication of
pressure, and the electrical signal from the transducer is carried to receiving instruments
by two wires. Electrical power is also required for the operation of the transducer.

A variation of the strain gauge pressure transduce r uses bonded single crystal
semiconductor wafers, whose capacitance varies with strain or distortion. Exhibit 1 I .4
shows an application of the bridge principle in an electromechanical pressure sensor that
uses capacitors as bridge elements . Here, the sensing element is a diaphragm exposed
to the pressure medium. When the diaphragm is deflected by applied pressure, the
capacitance of the element changes in proportion to the pressure. The measuring
element is a stiff diaphragm whose slight motion is proportional to the differential
pressure across it, with the capacitance circuit detecting that motion.

11.1.4 Selection of Pressure Measuring Instruments

The selection of a pressure measuring device for steam depends upon a number of
factors including range, accuracy, sensitivity, size, and cost. Some factors which
should be considered when selecting a pressure transducer suitable for a particular
application are discussed below.

Range:The of pressure transducer chosen will depend upon the expected

range of pressure to be measured; and whether the pressure to be measured is
absolute, differential, or both.

Ouality of Measurement: Choice of pressure transducer will depend upon the

desired accuracy of the transducer; the desired sensitivity of the transducer
(sensitivity is the ratio of change in transducer output to a change in the value of
the measured); the desired speed of response of the transducer; the desired
linearity of the transducer (linearity is defined as the closeness of a calibration
curve to a specified straight line); and susceptibility to hysteresis (hysteresis is
defined as the difference in output for a pressure value when the pressure value
is approached with increasing pressure and then with decreasing pressure).

Installation: The type of pressure transducer chosen will depend upon the
physical conditions of the desired location of the transducer (vibration, corrosive
environment, temperature, etc.); the desired location of the transducer readout
(e.g. control panel, site); and operational constraints (whether installation or
maintenance will effect the process).

Energy Conservation & Efficiency Project (ECEP)


 




Exhibit 11.3
Strain Gauge Pressure Transducer

 




Exhibit 11.4

 




INSTRUMENTATION FOR STEAM MEASUREMENT 11-8

C ost: The cost of a pressure transducer will depend upon the capital cost
incurred in the procurement and installation of the transducer; and the cost of
operating and maintaining the transducer.

11.2 FLOW METERS

A flow meter is a device that measures the rate of flow or the quantity of a moving
fluid in a conduit. It usually consists of both a primary device (sensor), and a
secondary device (readout and/o r totalizer). In this section, two commonly used types
of instruments for steam flow rate metering are described: orifice plates an d pitot
tubes.

11.2.1 Orifice Plates

Orifice plates are commonly used to measure steam flow. They are easy to fabricate
because of their simple design -- a metal plate with an accurately machined hole. In
most applications they are inexpensive and installation is the major cost. They provide
accurate and reliable data for flow measurement, particularly for medium flow rates in
pipes.

Orifice plates do not directly measure flow but instead measure a drop in pressure that
is proportional to flow rate. The operating principle of orifice plates is based on the
relationship between velocity and pressure of a flowing fluid. When a restriction, such
as a small diameter orifice plate, is inserted in a stream, fluid velocity must increase
when passing through it. The rise in velocity is accompanied by a proportional drop in
pressure (Exhibit 11.5).

In practical terms, this relationship is valid onl y when certain operation and installation
guidelines are followed: there is a minimum straight length of pipe upstream and
downstream from the orifice plate (Exhibit 11.6); pressure taps for measuring pressure
drop are a certain distance upstream and downstream of the orifice plate; the pressure
taps are clean and unobstructed; and the flow rate is above a minimum. The

drop in pressure, dp, caused by the orifice can be measured in a number of ways,
discussed later in the chapter.

There are a number of orifice plate designs with the primary differences amon g these
being the shape of the orifice. For the majority of energy flow measurements involving
steam, the square-edged orifice plate is recommended. The leading face of th e square-
edged orifice is flat and the edge of the orifice is sharp and square (Exhibit 11.7). The
orifice is sized for a particular range of flow rates. Exhibit 11.8 shows several typical
orifice plate installations.

There are four common methods of installing orifice plates . Using orifice flanges, the
pressure taps are drilled into the flanges themselves which are either welded or screwed
onto the pipe. The orifice is inserted and secured between them. Usin g carrier rings,
pressure taps are drilled into the rings and the orifice is inserted between them. The

Energy Conservation & Efficiency Project (ECEP)


 




High Pressure Low Pressure

Exhibit 11.5
Flow Through an Orifice Plate

 




250 4D

300

Exhibit 11. 6
Minimum Straight Length Piping for
Orifice Plates and Pitot Tubes

 




Exhibit 11.7
Concentric Square-edged Orifice

 




Exhibit 11.8 ’
Typical Installations for Orifice Plates

 




INSTRUMENTATION FOR STEAM MEASUREMENT 11-13

entire assembly is inserted between existing pipe flanges. With existing flanges, the
plate can be inserted between existing pipe flanges and pressure taps drilled in the pipe
itself. Although widely used in the past, this method has been replaced wit h more
standardized and convenient orifice flanges with flange taps. Prefabricated meter runs
consisting of a piece of pipe cut to a specified length with an orific e inserted and
pressure taps drilled can be installed in the pipe run. This is particularly useful in
diameters below 50 mm, where pipe roughness may prohibit the use of standard orifice
plates.

A variety of pressure tap configurations are available for orifice plates . Exhibit 11.9
illustrates each of the existing tap options. Flange taps are the most commonly
recommended configuration since they are convenient to use and can be easily replaced.
Taps are drilled in special orifice flanges 25 mm upstream and downstream of the plate
as shown in Exhibit 11.9(a). Comer taps are drilled into a carrier ring that can be
inserted between existing flanges, as in Exhibit 11.9(b) . Radius, vena contracta, and
pipe taps, although widely used in the past, have bee n largely replaced by standardized
flange taps. These taps are drilled directly into the pipe, and the orifice located
between existing flanges. Radius taps (c) are one diameter upstream and one half
diameter downstream. Vena contracta taps (d) are two and one half diameters
upstream, and the downstream tap location depends on the beta ratio. Pipe taps (e) are
located two and one half diameters upstream and eight diameters downstream.

To measure the pressure differential, pressure tubing is connected to the taps and run to
a device that measures differential pressure. These tubes are usually filled with the
fluid being measured. Normal installation requires that they be slanted to prevent
entrapment of gas bubbles if liquids are measured, or liquid columns if gas is measured.
In case of condensable fluids such as steam, condensate is allowed to fill tubes. If only
small amounts of condensible are present in the gas, these are collected and drained
periodically.

There are four basic types of equipment available to read dp and convert it to flow rate.
These are manometers, differential pressure gauges, differential pressure cells, and
microprocessors. Both portable an d permanently mounted versions of these instruments
are commonly used

11.2.2 Pitot Tubes

Pitot tubes are used to measure gas flow in ducts and pipes. Generally , pitot tubes are
less than 12.5 mm in diameter. They are inserte d into the stream at right angles to the
flow, and measure the velocity at a single point only. Pitot tubes are especially suited
for low to medium flows in large ducts . Pitot tubes tend to be less expensive than other
meters such as venturis or orifices. Because of their small size, pitot tubes cause very
little permanent pressure loss and are also appropriate for applications where a very low
permanent pressure drop is allowable . They are very attractive where a permanently
mounted device is not necessary, since they can b e inserted for spot checks and then
withdrawn.

Energy Conservation & Efficiency Project (ECEP)


 




(a) Flange Taps

(b) Carrier Rin g With Corner Taps

(c) Radius Taps

5/2-l/2 0 Variable: pt. Of greatest A D

( d ) Ven a Contract a Taps

Exhibit 11.9
Pressure Tap Configuration

 




INSTRUMENTATION FOR STEAM MEASUREMENT 11-15

Pitot tubes, like orifice plates, measure a differential pressure that relates to the flow
rate. Unlike orifice plates or venturi tubes, flow does not pass through the pitot tube.
Instead, flow impacts on the pitot and causes a higher pressure at the impact point . The
difference between this higher impact pressure (dynamic pressure) and the static
pressure of the fluid relates to the velocity of the fluid. Typical pitot tube designs are
shown in Exhibit 11.10. It is important to note that a pitot tube infers velocity at a
single point only. By determining velocity at different points, an average velocity can
be found from which an average flow rate can be calculated.

The operation of a pitot tube is based on the principle that when a flowing fluid impacts
an object, its velocity drops to zero and pressure at impact increases. This impact
pressure is greater than the free stream or static pressure measured at the wall. The
difference between these pressures is related to flow velocity and flow rate.

As for orifice plates, this relationship in practical terms is valid only when certain
operation and installation guidelines are followed. These guidelines include accurate
determination of average velocity, square impact of flow on the impact tap, cleanliness
of the taps, and proper flow ranges.

The major design differences among pitot tubes are the placement of static and dynamic
pressure taps, and whether the tube is permanently mounted or portable. The position
of pressure taps is more a concern of convenience than accuracy or application. With
some designs, these taps are separate, with the static tap located at the duct wall and the
impact tap at the end of the pitot Other designs put both taps on the tube, and some
use several impact taps to yield an average pressure difference. Any of these designs
can be permanently installed in the flow duct. Some of the more common types of pitot
tubes are described below.

Permanent Mounting With Separate Static Tarps: Thispitot tube impact tap
is permanently mounted facing directly upstream . It is located at a point where
the velocity measured represents average velocity or the reading is corrected by
ap at other points. The static tap is drilled through the wall of the pipe not more
than one half diameter upstream of the impact tap. Both taps are screw fitted
into place as seen in Exhibit 11.10(a).

Averaging Pitot Tube (Annubar): This is a diamond shaped strut that spans the pipe
as shown in Exhibit 11.10(d). There is more than one impact hole

along its length on the upstream side of the probe and a single downstream rear
facing static pressure port. This arrangement gives an average velocity across
the pipe.

Portable Pitot Tubes With Stati cTaps: For portable pitots, the impact tap and static
tap are both located on the tube as in Exhibit 11.10(b). The impact tap faces squarely
into the flow. The static tap is usually located on the circumference of the impact tube
and may be located on the downstream side of the impact tube or at right angles to the
flow. The tube can be permanently mounted with screw fittings, as in Exhibit
11.10(c), but is usually used as a portable instrument. Portable pitot tubes are
especially useful in measuring

Energy Conservation & Efficiency Project (ECEP)


 




(a) Single Hole pitot tube ( b ) P o r t a b l e Pitot T u b e

Separate Static Tapping

(c) Commercial Pitot Tube ( d ) Averaging Pitot Tube ( A n n u b a r )

Exhibit 11.10
Tube Design and Installation

 




INSTRUMENTATION FOR STEAM MEASUREMENT 11-17

exhaust or intake flows from open- ducts. The measurement can bc made very
quickly without disrupting the process.

Pitot tubes are appropriate for measuring steam flow in round pipes with diameters
greater than 100 mm. especially when venturis or orifices are too exoensive or cause
too great a permanent pressure ‘drop. The pitot tube must bc used in fully developed
flow by assuring minimum straight lengths of pipe upstream and downstream of the
tube. Permanent mountings are particularly attractive when repeated readings are to be
made but other meters are too expensive.

Placement of the pitot tube is critical. The pitot tube measures velocity at only one
point, but this velocity must represent the average velocity for the flow equation to be
valid. In all ducts velocity varies with position pcrpcndicular to flow. To find the
point of average velocity, the duct is divided into discrete s ections and velocity is
measured at several points in order to calculate the average. The pitot tube can then be
placed at the point of average velocity for extended m easurements.

To find flow rate, differential pressure must be measured and co nverted to mass flow.
Pitot tubes are essentially special pressure taps, so no other tap is needed. The pressure
signal must be transferred, however, to a dp measu ring device; rubber tubing is
typically used. The three basic methods for measuring dp are man ometers, dp cells,
and velometers. Velometers are mechanical pressure gauges that sense the difference
between the high and low pressure taps. Rather than indicating the value of pressure
difference, velometers are calibrated and graduated to give a direct reading of velocity.
An instantaneous value of velocity can be read directly from the velometer, and flow
can be determined from the flow equation, calibration curves, or tables.

11.2.3 Selection and Application of Flow Meters

The selection of flow meter for metering a particular fluid depends upon a number of
factors including the temperature, pressure, and expected flow rate of the steam.

Some factors that should considered when selecting a flow meter are discussed below:

Steam Characteristics: Selection of a flow meter will depend on the

characteristics of the steam: expected temperature range of the steam; expected
pressure range of the steam; and expected flow rate of the steam.

Quality of Measurement: Selection of a particular flow meter will depend

upon: desired accuracy of the meter; and desired flow rate range of the meter.

Installation: Installation constraints to be considered when selecting a flow

meter for a particular application may be: operational constraints (whether
maintenance or replacement of the flow meter will affect the process); and the
pipe size in which the flow meter is to be installed.

Energy Conservation & Efficiency Project (ECEP)


 



INSTRUMENTATION FOR STEAM MEASUREMENT 11-18

Cost: The cost of a flow meter will depend upon: the typical cost incurred in
the procurement and installation of the flow meter; and the cost of operation
(permanent pressure loss in fluid flow) and maintenance of the flow meter.

 



APPENDIX A STEAM TABLES

Energy Conservation & Efficiency Project (ECEP)