Topic 4: PIPING

Learning Outcomes
Upon successful completion of this topic you will be able to:
• determine the layout and sizing of the compressed air piping system
including consideration of frictional pressure drop and condensate removal

Introduction to the Topic

Mechanical, civil and chemical engineers frequently have to design piping or
ducting systems to distribute steam, water, air etc. This topic deals with many of
the issues involved in designing piping.

Background Skills and Knowledge

Students will require the following:
• Familiarity with the calculation of pipe friction losses.

Activity 4A - Reading

Read all of Chapter 4 below i.e. Section 4.1 “General requirements in piping
design”; Section 4.2 “Brief review of compressed air distribution systems” ;
section 4.3 “Pipe internal diameter”; and Section 4.4 “Pipe thickness and
supports”

School of Engineering
Page 1 of 31
Chapter 4: Piping
Section 4.1 General requirements in piping design
Mechanical and civil engineers are frequently required to design piping systems,
the general objective being to deliver a fluid to the locations where it is needed.
Typical examples include:
• Water for industrial, commercial and domestic use
• Waste water removal and storm water drainage
• Steam in power stations and factories
• Compressed air in factories
• Hot water and chilled water for building heating and cooling
• Refrigerant in air conditioners and refrigerators
• Fuel oil and gas in industry, commerce, homes and vehicles.
Issues that the engineer must address include:

(a) Deciding on pipe internal diameter. This must be:

(i) Large enough to pass the required flow without excessive frictional pressure
drop because large pressure drops mean wasted power and hence running
cost penalties.
(ii) Not so large that capital cost is being wasted on the pipe material, and the
contained fluid if that is expensive, as is the case for refrigerants.
(iii) Not so small that velocity is high enough to create nuisance noise or
erosion.
(iv) Of such a size that any secondary fluids present are either dropped out
of the flow if that is desired, or carried along if that is desired. Examples of
secondary fluids are: air bubbles in a water system, condensate in a steam
system, condensate and oil in a compressed air system, oil in a refrigerant
system (this lubricating oil must be returned to the refrigerant vapour
compressor via the refrigerant pipes if possible, or separated and returned
by another route).

(b) Deciding on the pipe material, and associated joining methods and
materials, which must:
(i) Be compatible with the fluid and not corroded too quickly.
(ii) Have reasonable strength at the operating temperature – not adversely
weakened by high temperature or embrittled by low temperature.
(iii) Be of reasonable first cost.
(iv) Be able to be fabricated at reasonable cost (e.g. threaded, flared or welded
joints can be made; bends can be made or fitted; etc.)

School of Engineering
Page 2 of 31
 (c) Deciding on the pipe thickness which must be:
(i) Thick enough to safely carry the stresses caused by the fluid pressure,
including “water hammer” if that is present. Crowe et al. (2009, p.187- .
192). 1 show how to calculate water hammer pressure rise. Devices are
available which reduce water hammer or alternatively relieve the excess
pressure.
(ii) Thick enough to safely carry the extra stresses caused by other loads such
as: pipe weight and fluid weight particularly where the pipe is only
supported at intervals, inertia loads where the fluid flow changes
direction, pressure pulses caused by the fluid flow being stopped
suddenly, and thrusts or pulls caused by thermal expansion or
contraction.
(iii) Thick enough to allow for any corrosion and/or erosion that will occur
over the life of the pipe.
(iv) Thick enough to have threads cut in it, or to be flared or to be welded as
the case may be.
(v) Not excessively thick as this represents waste of capital and resources.

(d) Deciding on the appropriate location for the pipe which must:
(i) Be convenient to the user.
(ii) Allow for future connections and extensions if likely.
(iii) Allow for drainage facilities if necessary.
(iv) Be accessible for maintenance purpose.
(v) Not take an excessively long route as this will both increase capital costs
due to the extra material used, and increase running costs due to the extra
frictional pressure drop.
(vi) Not be exposed to impact.
(vii) If possible, not be exposed to temperatures which could freeze its
contents. If the pipe does contain water or other fluids liable to freezing,
and has to be outside then it should be heavily insulated, fitted with
heating arrangements of some sort, buried or drained in freezing
conditions.
(viii) Protected from rain if this is liable to corrode it or damage
insulation.

1Crowe CT., Elgar D.F., Williams B.C., Roberson J.A., 2009, Engineering Fluid mechanics, 9th ed., J. Wiley &
Sons., New York, p.187-192.

School of Engineering
Page 3 of 31
 (e) Deciding on the method and frequency of supports, including allowance
for expansion/contraction.

(f) Deciding if the pipe should have a slope.

(g) Deciding if expansion joints are necessary for the pipe material, and
expansion vessels for the fluid.

(h) Deciding if thermal insulation is required and if so, what is the economic
thickness.
Hot water and steam pipes, and chilled water and refrigerant pipes are
usually insulated. The economic thickness is determined by considering the
capital cost of the insulation and the running cost over the life of the
installation of the lost heat or “coolth”.

(i) Deciding on what control and processing fittings should be incorporated,

such as:
isolating valves, flow regulating valves, non-return valves, safety valves,
pressure regulating valves, drain valves, air vents, filters, secondary fluid
separators, lubricators, flow meters and switches, pressure gauges and
switches, temperature gauges and switches, sight glasses, inspection
openings, traps, etc.

Advice on these various pipe sizing, routing, etc. decisions can often be found in
literature specific to the particular application. For example, the American Society
of Heating, Refrigeration and Air-conditioning Engines (ASHRAE) publishes
various “Handbooks” and the Australian Institute of Refrigeration, Air-
conditioning and Heating (AIRAH) publishes various “Application Manuals”
which give advice on the sorts of piping applicable to building heating and cooling,
and refrigeration. The following references give advice applicable to compressed air
piping systems:
• Barber, A., 1997, Pneumatic Handbook, 8th ed., Elsevier Science, New
York.
• Rollins, J.P. (editor), 1973, Compressed Air and Gas Handbook , 4th
edition, Compressed Air and Gas Institute, U.S.A.
• Loomis, A.W. (editor), 1980, Compressed Air and Gas Data, 3rd edition,
Ingersoll-Rand Company, U.S.A.
• Spirax Sarco Limited, 1978, Compressed Air, Spirax Sarco Limited, U.K.

Some of the decisions in the prior list will be influenced by government regulations
and Standards (e.g. “AS4041, Pressure Piping”, published by Standards Australia.)
particularly where safety and hygiene are involved. The remainder of this chapter
will be a brief presentation of material relevant to compressed air systems. The
material that will be presented here is not exhaustive and the reader is referred to
the literature for a fuller coverage.
School of Engineering
Page 4 of 31
Section 4.2 Brief review of compressed air distribution systems
4-2-1 Mains route
As with most piping supply systems, compressed air factory systems have a main
pipe, or pipes, through the factory from which branch lines are run to various
workshops, etc. Two basic mains systems are used, as shown in plan view in Figure
4.2a and Figure 4.2b.

Fig. 4.2a: Dead End Main Fig. 4.2b: Dead-end Split-flow Main
© RMIT University, 2013 (Dixon C,.,Kissane M.)

Fig. 4.2c: Ring Main © RMIT University, 2013 (Dixon C,.,Kissane M.)

Some factory layouts particularly lend themselves to the ring main system which
has some claimed advantages: firstly that plant at the far end of the factory has two
routes by which air can reach it so the pressure drop may be less [although if the
alternative dead-end main is adequately sized, this claimed advantage is debatable],
and secondly that the ring main has appreciable volume and can therefore act as
storage, like the receiver, assisting to minimise drops in pressure when brief sudden
demands are made. Obviously, the ring main having a greater amount of pipe in it,
will have a higher first cost, and more potential for leaks. The claimed running
advantages however may outweigh these disadvantages.

In Figures 4.2a and 4.2b a receiver is shown as being with the compressor and this
is usually the case. However if there is a plant item in the factory, far from the
School of Engineering
Page 5 of 31
 compressor, requiring to be fed large intermittent flows, then an additional
appropriately sized and located receiver may be advantageous. It is left to the
student to consider where it should be located and what the advantage of such a
location would be. There is some analogy here with the secondary reservoirs used
in the city water supply systems discussed in Chapter 1.

Similarly Figures 4.2a and 4.2b show the compressor/s in one location, but there are
some advantages in having compressors at several locations around the factory. It
is left to the student to contemplate the advantages and disadvantages of this
alternative strategy.

4-2-2 Condensation in the mains

As was explained in the psychrometry of compressed air section of Chapter 2,
compressing the air, which has water vapour in it, raises the partial pressure of this
steam and hence the saturation temperature and dew point. The resulting dew point
is frequently above the ambient temperature, so cooling of the air in the pipes by
heat transfer to the ambient air can cause condensation to form in the pipes. This
water can cause all sorts of problems with tools and production processes, as was
described in Chapter 2. After-coolers are therefore commonly used to cool the air
leaving the compressor, and hence remove a lot of the moisture before the air goes
to the receiver and out into the distribution pipework. If moisture is a particular
problem, air dryers, such as were viewed in Activity 2G can be fitted at extra cost.
Since it is likely, in the common situation where a dryer is not used, that some
moisture will still condense in the mains, these mains must be arranged to allow for
its presence. Similarly in steam piping arrangements must be made to cope with
condensate. Therefore for both steam and compressed air piping, it is good practice
to:
(i) Make take off connections from the top of the main to minimise the likelihood
of carryover of condensate, scale and grit. Such branch connections should of
course have an isolating valve.

(ii) Arrange fittings in the main at regular intervals that can separate the
condensate from the air (or steam). At its simplest, this could be a tee as shown
in Figure 4.2d, noting that the branch should not be smaller than the main run,
i.e. not as in Figure 4.2e

School of Engineering
Page 6 of 31
 Fig. 4.2d Better drain tee © RMIT University, 2013 (Dixon C,. Marchiori G. Kissane M.)

Fig. 4.2e Poor drain tee © RMIT University, 2013 (Dixon C,. Marchiori G. Kissane M.)

An alternative arrangement, also using common pipe fittings that should give
superior separation is that shown in Figure 4.2f, and it is frequently recommended.
In addition to using gravity to separate the water from the air this arrangement
involve momentum principles, with the water droplets being flung out of the air
School of Engineering
Page 7 of 31
flow because the flow changes direction. This arrangement will unfortunately have
a higher frictional pressure drop than the straight through tee.

Fig. 4.2f Superior drain tee © RMIT University, 2013 (Dixon C,. Marchiori G. Kissane M.)

School of Engineering
Page 8 of 31
 Activity 4B - Commercial water separators
Commercial water separators using this combination of gravity and
momentum separation principles are also available, such as those examples
described below
Step 1: Please refer to the Spirax separators (PDF 106kB) 2
Step 2: Please refer to the information on the workings, fitting and efficiency
of Separators (PDF 229kB) 3
Step 3: In some commercial separators the air and water are fed into a small
vessel tangentially, and by centrifugal action the water is thrown onto the wall
of the vessel from which it then drains down by gravity.
Please refer to the e-reserve document detailing separator arrangements and
associated fittings (PDF 1.5MB) 4

These water separation devices, such as the tee in Figure 4.2f, should be fitted at
regular intervals to deal with the condensate that has precipitated in the pipe since
the air/steam passed the previous separation device. The recommended maximum
spacing between such devices in compressed air lines is 30m (Spirax Sarco Limited,
1978, p.11) 5 The dead-ends of mains and branch lines should also be drained.
Similarly a branch line should have a water separating device fitted before the air
or steam goes off to its final point of use.
Fig 4.2g shows a float operated “trap” that automatically empties condensate that
has built up in the drains legs from separation devices. Other air/steam processing
devices may also need to be included before the fluid goes into the tool or process,
such as filters, lubricators and pressure regulators.

2 Spirax Sarco Limited, 1978, Separators, Compressed Air: How to make the most of your system from

compressor to point of use. Practical Study: PS12, viewed on 28th August 2013 < https://ap01-
a.alma.exlibrisgroup.com/view/delivery/61RMIT_INST/12145244200001341 >

3 Spirax Sarco Limited, 1982, Separators, Compressed Air: How to make the most of your system from

compressor to point of use. Practical Study: PS12, viewed on 28th August 2013 < https://ap01-
a.alma.exlibrisgroup.com/view/delivery/61RMIT_INST/12164690170001341>

4 Kaeser Compressors Australia, 2010, Condensate: correct drainage, Compressed Air Engineering: Basic

principles, tips and suggestions (P-2010ED.5/06) viewed on 28th August 2013 <https://ap01-
a.alma.exlibrisgroup.com/view/delivery/61RMIT_INST/12164690190001341>

5 Spirax Sarco Limited, 1978, The Distribution System, Compressed Air: How to make the most of your system

from compressor to point of use. Practical Study: PS12, p.11

School of Engineering
Page 9 of 31
 Fig. 4.2g Schematic of float type drain trap © RMIT University, 2013 (Dixon C, Marchiori G.)

Activity 4C - Commercial float trap

Please refer to the sectioned view of a commercially available float trap which
may be seen on page 231 of the 8th edition of the Pneumatic Handbook (PDF
48KB) 6

Continuing with, for both steam and compressed air piping, it is good practice to:

(iii) Slope the main slightly downwards in the direction of flow so that
condensate moves steadily towards the drains helped by gravity and the air
flow. The recommended minimum fall for compressed air lines is 1 in 100 or 1
in 120. (Barber 1997, p.221). 7 Having the pipe horizontal with infrequent drains,

6Barber, A., 1997, Water drain traps, Pneumatic Handbook, 8th ed., Elsevier Science, New York, p.231, viewed
on 28th August 2013 <https://ap01-
a.alma.exlibrisgroup.com/view/delivery/61RMIT_INST/12187811850001341>

7 Barber, A., 1997, Pneumatic Handbook, 8th ed., Elsevier Science, New York, p.221

School of Engineering
Page 10 of 31
 or worse still sagged, may lead to a build-up of condensate in the pipe and then
formation of slugs as shown in Figure 4.2h. Such slugs can destructively
“hammer” into valves etc if carried forward by high speed steam or air flows.

Fig. 4.2h Condensate slug formation in poorly drained compressed air or steam pipe.
© RMIT University, 2013 (Dixon C,. Marchiori G. Kissane M.)

(iv) Have the pipe diameter sufficiently large to ensure the air/steam is going
sufficiently slowly such that it does not pick up and carry along (“entrain”) the
condensate. For compressed air systems it is suggested by the British
Compressed Air Society that the velocity not exceed 6m/s to avoid carry over of
moisture past the drain legs (Barber 1997, p.219). 8 on the basis of avoiding

8 Barber, A., 1997, Pneumatic Handbook, 8th ed., Elsevier Science, New York, p.219

School of Engineering
Page 11 of 31
 moisture carryover past the drain legs. It has already been noted that the
recommended maximum spacing between water separation devices is 30 m.
Barber further suggests that for branch lines that do not exceed 15 m this velocity
limit may be increased to 15 m/sec.

Section 4.3 Pipe internal diameter

4-3-1 Compromises required in deciding internal diameter
Decision on the pipe internal diameter to be used often involves a compromise
between conflicting aims. Having the pipe smaller obviously lowers first cost, but
because the fluid (e.g. water, steam, air, refrigerant) moves faster in smaller pipes,
the frictional pressure drop is greater, and therefore the running cost will probably
be greater. For example in a compressed air system where a pressure of, say, 690
kPag is required at the outlets for the tools to work properly, the compressor has to
deliver a pressure sufficiently above this to compensate for the pressure loss due to
pipe friction; therefore, if the pipe friction loss is greater, the compressor will have
to work harder and hence consume more power and so make the running costs
greater. Similar arguments apply for other fluid systems.

There is thus a compromise to be had between the capital cost of the pipe, which
decreases with smaller pipes, and the running costs which increases. Economically
the best size would be that which gives the lowest total cost over the life of the
system, and this would obviously be affected by the relative cost of materials and
power, which may vary with location.

Rather than doing a full economic analysis each time a pipe is sized, approximate
guidelines on economically acceptable pressure drops have been developed for
various fluid applications. In the case of compressed air, for example:

School of Engineering
Page 12 of 31
 (a) Loomis (“Compressed Air and Gas Data” ) suggests that for a system
with 690 kPag at the compressor/receiver: (Loomis 1980 p.27.6, 27.8). 9
(i) The pressure drop in the main lines, including fittings, should be less than
20 kPa based on the average rate of flow, which is what the flow tends to
be in main line.
(ii) Branch mains do not have as great a diversification of demand as main
lines and must therefore be sized for their maximum flow. This may run
150 to 175 percent of the average, depending on the number of outlets to
be in use at any one time. Pressure losses in this section should also be
less than 20 kPa.
(iii) Feed lines from the branch main to the floor should be sized for the
maximum flow of as many tools as may be connected at any time. These
lines are usually short and pressure drop should be less than 7 kPa.

This gives a total pressure drop from receiver to most distant tool of (20 + 20
+ 7 =) 47 kPa or 7% of the pressure (gauge) at the compressor/receiver.

(b) Barber suggests: “Limit the pressure drop to the point of use to a
prescribed value e.g. 0.1 bar or 0.5 bar or as a proportion of the supply
pressure, say 5%”. (Barber 1997, p.220). 10

For the purposes of the assignments associated with this subject, the
frictional pressure drop along the main line from the outlet of the tee at the
receiver to where the most distant branch line starts is not to exceed 20 kPa.

The higher velocity that occurs in smaller pipes, as well as causing greater
frictional pressure drop, may have other detrimental effects such as noise,
erosion, high pulse pressures when valves are shut suddenly, and
entrainment of secondary fluids where that is not desired. (Note however
that in some circumstances entrainment may be desired, e.g.: to avoid
bubbles forming air locks in water pipe systems at high spots if the air can’t
be vented there; or for the purpose of returning oil along the cold suction
pipe to a refrigeration compressor).

9Loomis, A.W. (editor), 1980, Compressed Air and Gas Data, 3rd ed., Ingersoll-Rand Company, U.S.A.
p27.6,27.8

10 Barber, A., 1997, Pneumatic Handbook, 8th ed., Elsevier Science, New York, p.220

School of Engineering
Page 13 of 31
 For the case of compressed air it has been suggested that in the main lines
the velocity should not exceed 6 m/s so that the condensate is not entrained
and will separate out at the separation devices (which should be no more
than 30m apart). For short branch lines, less than 15m long, an upper limit of
15 meters per second has been suggested. (Barber 1997, p.219). 11

Barber’s suggested design approach to size the main line is:

• Firstly, using the velocity limit just quoted find commercially available
“trial” pipe sizes to use along the route, that keep the velocity below the limit.
• Secondly, using these trial pipe sizes determine the pressure drop and check
that it is not excessive.
• If the pressure drop is excessive, try larger internal diameters and repeat the
pressure drop calculation.

Obviously, as branches come off the main less air will be flowing in the remainder
of the main, so the diameter will reduce in stages along the route.

4-3-2 Calculation of fluid velocity and frictional pressure drop in pipes

(a) Velocity
Students will recall from their fluid mechanics studies that the mass flow rate
and volume flow rate in a pipe are related to the average velocity in the pipe
via the following three equations, which can be used to determine the
velocity associated with a given mass or volume flow:

m = ρ Vr Eqn. PI1

Vr = CA Eqn. PI2

∴ m = ρ CA Eqn. PI3

where Vr is the “restrained” (not free) volume flow rate [m3/sec]
C is the average velocity (aka celerity) [m/sec]
Π 2
A is the cross sectional flow area D [m2]
4

11 Barber, A., 1997, Pneumatic Handbook, 8th ed., Elsevier Science, New York, p.219

School of Engineering
Page 14 of 31
 ρ is the fluid density [kg/m3]
Students will recall from their thermodynamic studies that if the fluid is a vapour, the
density ρ must be obtained from thermodynamic tables, but if it is a gas and can be
approximately treated as a perfect gas, then ρ may be obtained via the perfect gas
equation, e.g. for air:

pV = mRa T Eqn. PI4

m p
∴ ρ= = Eqn. PI5
V Ra T

where Ra is the gas constant for air [287 J/(kg.K)]

T is the absolute temperate [K]
p is the absolute pressure [N/m2]

(b) Frictional pressure drop

Consider now an elemental control volume of length ∆x in a pipe, as shown
by the dotted boundaries in Figure 4.3a. The pipe centreline does not change
in height so gravitational effects are not changing, the internal diameter is
constant, and the mass flow rate m is steady.

Fig. 4.3a An elemental control volume in a pipe,

© RMIT University, 2013 (Dixon C,. Marchiori G. Kissane M.)

School of Engineering
Page 15 of 31
 L is the pipe length [m]
D is the internal diameter of the pipe [m]
τ is the shear stress at the wall [N/m2]
f C2
and τ = ρ
4 2

where f is the Darcy-Weisbach resistance co-efficient or friction

factor, which is dimensionless and is a function of the
relative roughness of the internal pipe surface and the
Reynold’s Number

The shear force Fτ acting on the fluid moving through the control volume is equal
to the shear stress multiplied by the surface area at the wall.
∴ Fτ = τ x ΠD x ∆x
f C2
Fτ = ρ ΠD∆x
4 2
The net force acting on the system is equal to the net efflux of momentum from the
control volume.

Π 2 Π f C2
∴ px D − ( p + ∆p ) x D 2 − ρ ΠD∆x = m [(− C ) + C + ∆C ]
4 4 4 2
Π 2
and m = ρ CA = ρC D
4
Π 2 f C2 Π 2
∴ −∆ p D − ρ ΠD∆x = ρC D [∆C ]
4 4 2 4
Π 2
∴ Dividing through by D
4
∆x C 2
⇒ ∆p + f ρ + ρ C∆C = 0 Eqn. PI6
D 2

Since m = ρ CA , and A is fixed, if the density does not change the velocity C will
not change, i.e. ∆C = 0 , and hence equation PI6 becomes:

∆x C 2
∆p = − f ρ
D 2

∆x C 2
or frictional pressure drop ∆p f = f ρ
D 2

School of Engineering
Page 16 of 31
So for a pipe of length L, in which an incompressible fluid at constant temperature,
i.e. of fixed density, flows at a steady velocity:

L C2
∆p f = f ρ Eqn. PI7
D 2
which is the well-known Darcy-Weisbach equation.

Some engineering pipe flow designs however involve fluids that:

• Are not incompressible, and when the pressure drops the density decreases,
and hence the velocity must increase if the flow area is fixed and the mass
flow is continuing steadily (m = ρCA) . This velocity increase will cause more
pressure drop than the simple Darcy-Weisbach formula would indicate. This
increase in pressure drop can be seen by considering equation PI6.
• Involve two phases such as steam and condensate.
• May even involve a third “component”, such as oil in a refrigerant liquid
vapour mixture.
• Are moving at close to the speed of sound and subject to the effects of shock
waves and choking.

More sophisticated equations, such as result from solving the full differential
equation PI6, are available in the literature, however for compressible single phase
gases moving at speeds significantly less than the speed of sound, the Darcy-
Weisbach equation may be used, but not indiscriminately, subject to the limitations
outlined below (Loomis 1980 p.26-8). 12
1. ∆p less than 10 percent of p1 – reasonable accuracy will be obtained if ρ and
C are based on upstream or downstream conditions, whichever are known;
2. ∆p is between 10 and 40 percent of p1 – the Darcy equation will give
reasonable accuracy by using ρ and C based on the average upstream and
downstream conditions ( p1 + p 2 ) / 2 . Several trials may be required; and,
3. ∆p greater than 40 percent of p1 – the Darcy formula cannot be applied in a
single step. It is necessary to divide the total length into shorter sections in
which each ∆p will be less than 40 percent of p1 (per section). Then add the
sectional drops to obtain the total.”

12 Loomis, A.W. (editor), 1980, Compressed Air and Gas Data, 3rd ed., Ingersoll-Rand Company, U.S.A. p.26-8

School of Engineering
Page 17 of 31
(p1 and p2 are the absolute values of pressure at the beginning and end respectively
of the pipe length.)

For the particular case of compressed air piping in factories, the following further
comments are made:
(i) If the frictional pressure drop is approaching 40% of the upstream pressure
there is a fault in the design (e.g. the pipe diameter is too small) and this
should be remedied.

(ii) Velocities of 6 m/sec and 15 m/sec are much less then the speed of sound.
For an ideal gas the speed of sound a is found from:

a = γRT

where γ is the ratio of specific heat at constant pressure

to that at constant volume
[ ≈ 1.4 for air]

R is the gas constant [=287 J/(kg.K) for air]

Thus for air at 27°C [300K]

a = 1.4 x 287 x 300 = 347 m/sec

(iii) The friction factor f depends, as stated earlier, on the relative roughness
of the pipe and the Reynold’s Number [Re].

ρ CD
Re = Eqn. PI8
µ
where µ is the absolute or dynamic viscosity of the fluid
[N.sec/m2=Pa.sec]

Recalling that
m
m = ρ CA ∴ ρC =
A

m
D
⇒ Re = A Eqn. PI9
µ

School of Engineering
Page 18 of 31
 For gases µ is chiefly a function of temperature, and pressure does not
influence µ more than 10% within moderate ranges of pressure, i.e. for
pressure less than 3450 kPag (Loomis 1980, p. 26-3). 13
Therefore if the compressed air in the pipe is at a constant temperature along
the pipe, µ is constant. Furthermore if the mass flow and flow cross sectional
o
m
area are constant along a length of pipe, is constant. Hence for these
A
conditions it can be seen from equation PI9
m
D
Re = A
µ
that the Reynold’s Number in the length of pipe (of uniform diameter) is
constant; therefore the friction factor, which depends on Reynold’s Number
and relative roughness, will also be constant along the length of pipe,
assuming the roughness doesn’t change.
(iv) If a compressed air tool is percussive, “the flow required is pulsating and the
pressure pulsations will be felt as far back in the system as the nearest
reservoir of substantial size. The friction factor appropriate to pulsating flow
is greater than that calculated on the basis of the average flow requirement.
The general recommendation is to increase the factor by at least 50% to allow
for this”(Barber 1997, p.144). 14

(v) The effect of water vapour on the air mixture’s density and viscosity is small
and usually neglected in pressure drop calculations.

(vi) If the condensate does not build up in the pipe because it is drained off at
frequent intervals along the pipe its effect on pressure drop is small and
usually neglected.

Returning now to the Darcy-Weisbach equation:

L C2
∆p f = f ρ ,
D 2

13 Loomis, A.W. (editor), 1980, Compressed Air and Gas Data, 3rd ed., Ingersoll-Rand Company, U.S.A. p.26-3

14 Barber, A., 1997, Pneumatic Handbook, 8th ed., Elsevier Science, New York, p.144

School of Engineering
Page 19 of 31
f is the friction factor or resistance co-efficient and, as was stated earlier, is
dimensionless and a function of the relative roughness and the Reynold’s Number.
ε
The relative roughness = where ε is the absolute roughness of the internal pipe
D
which may be obtained from most standard fluid mechanics textbooks.

Activity 4D - Pipe roughness data

Step 1: Pipe roughness data

Please refer to the Pipe Roughness data found on page 3.7 of ASHRAE
Fundamentals (PDF 188KB). 15
ε
Note that if ε is in millimetres, then to obtain relative roughness , the
D
diameter D must be in millimetres for this particular small calculation.

Step 2: The internal diameter D

Deduce the internal diameter D from the pipe data in Table 7.2 of the Steel Pipe
Design Manual (PDF 72KB). 16
Note: The “nominal” bore is not the actual bore. The actual bore, i.e. internal
diameter, is to be used in flow calculations. The outside diameter given in the
above table is an actual dimension.

15Couvillion, R. J. (contributor), 2009, Conduit friction, ASHRAE handbook : fundamentals S I ed., American
Society of Heating, Refrigeration and Air-Conditioning Engineers, Atlanta, Ga. p. 3.6, 3.7, viewed on 28th
August 2013 <https://ap01-a.alma.exlibrisgroup.com/view/delivery/61RMIT_INST/12187811830001341>

16 Tubemakers of Australia Limited,1981, Table 7.2 Range of standard sizes tubeline pipe, Steel pipe design

manual, Tubemakers of Australia Limited Steel Pipe Division, Melbourne. p.76, viewed on 28th August 2013
< https://ap01-a.alma.exlibrisgroup.com/view/delivery/61RMIT_INST/12187201720001341>

School of Engineering
Page 20 of 31
Steel is not the only material used for compressed air lines. Other materials (Barber
1997, p.239-245). 17 include copper tube, ABS copolymer (Acrylonitrile – Butadiene
– Styrene), nylon (for small tools and for circuits in automation applications),
terylene braided PVC hoses (for tools) and textile reinforced rubber hose (for tools).

Activity 4E - Properties of air at atm pressure and friction factor

Step1: Properties of air at atm pressure

The dynamic viscosity µ , necessary for the determination of Reynold’s
Number, may be obtained from thermodynamic tables or charts.
Please refer to the Table A-15. Properties of air at atm pressure (PDF 124KB).
18

Note that, for perfect gases, pressure does not influence dynamic (aka
absolute) viscosity µ , :

Step2: Finding the ‘friction factor’

ε
Having obtained values of the relative roughness and the Reynold’s
D
Number, the friction factor is then found. This is often obtained from the
“Moody” diagram which may be obtained from most standard fluid
mechanics textbooks.
Please refer to the ’Moody’ diagram on page 3.7 of ASHRAE Fundamentals
(PDF 188KB). 19

17 Barber, A., 1997, Pneumatic Handbook, 8th ed., Elsevier Science, New York, p.239-245

18Çengel, Yunus A., Ghajar and Afshin J., 2011, Appendix 1, Heat and mass transfer : fundamentals &
applications, 4th ed., McGraw-Hill, New York, p. 884, viewed on 28th August 2013 < https://ap01-
a.alma.exlibrisgroup.com/view/delivery/61RMIT_INST/12141498760001341>

19 Couvillion, R. J. (contributor), 2009, Conduit friction, ASHRAE handbook : fundamentals S I ed., American

Society of Heating, Refrigeration and Air-Conditioning Engineers, Atlanta, Ga. p. 3.6, 3.7, viewed on 28th
August 2013 <https://ap01-a.alma.exlibrisgroup.com/view/delivery/61RMIT_INST/12187811830001341>

School of Engineering
Page 21 of 31
The friction factor may be calculated from the following expression, within the
ε
ranges of and Re indicated, to give a value that deviates no more than 3% from
D
those obtained from the Moody diagram (Crowe et al. 2009, p.331). 20

0.25
2
  ε 5.74  
log 10  3.7 D + Re 0.9  
f =   Eqn. PI10

ε
for 10 −5 < < 2 x 10 −2 and 4 x 10 3 < Re < 10 8
D

L C2
The equation ∆p f = f ρ is, as was stated earlier, for straight pipes. Where the
D 2
fluid flows through bends and other fittings such as valves, further frictional
pressure drop occurs. The pressure drop for each fitting is usually calculated either
of two ways. See Calculation 1 and 2 below.

C2
(i) Using the equation ∆p f = Kρ Eqn. PI11
2
where K is the loss co-efficient for the fitting.

Tables of loss coefficients are available for a range of fittings, and may be obtained
from most standard fluid mechanics textbooks.

20 Crowe CT., Elgar D.F., Williams B.C., Roberson J.A., 2009, Engineering Fluid mechanics, 9th ed., J. Wiley &

Sons., New York P.331

School of Engineering
Page 22 of 31
 Activity 4F - Calculating pressure drop for fittings

Calculation method 1:
C2
Using the equation ∆p f = Kρ Eqn. PI11
2
where K is the loss co-efficient for the fitting.

Tables of loss coefficients are available for a range of fittings, and may be
obtained from most standard fluid mechanics textbooks.

Please refer to Table 3 Fitting Loss Coefficients of Turbulent Flow on page 3.8
of ASHRAE Fundamentals SI ed. (PDF 223KB). 21

Calculation method 2:
Considering the fitting to be equivalent to a particular length of pipe, tabulated
values for which are available in various handbooks.

Please refer to Table 2 –Equivalent pipe lengths of valves and bends of the
Pneumatic handbook 8th ed. (PDF 26KB). 22

21 Couvillion, R. J. (contributor), 2009, Fluid Flow, ASHRAE handbook : fundamentals S I ed., American

Society of Heating, Refrigeration and Air-Conditioning Engineers, Atlanta, Ga. p. 3.8, 3.9, viewed on 28th
August 2013 <https://ap01-a.alma.exlibrisgroup.com/view/delivery/61RMIT_INST/12187811810001341>
22Barber, A., 1997, Pipe flow calculations , Pneumatic Handbook, 8th ed., Elsevier Science, New York, p.141,
viewed on 28th August 2013< https://ap01-
a.alma.exlibrisgroup.com/view/delivery/61RMIT_INST/12187811900001341>

School of Engineering
Page 23 of 31
Section 4.4 Pipe thickness and supports
As was outlined in the introduction to this chapter and will be repeated here for
convenience, pipe thickness has to be:
i. Thick enough to safely carry the stresses caused by the fluid pressure.
ii. Thick enough to safely carry the extra stresses caused by other loads such
as: pipe weight and fluid weight particularly where the pipe is only
supported at intervals, inertia loads where the fluid flow changes direction,
pressure pulses caused by the fluid flow being stopped suddenly, and
thrusts or pulls caused by thermal expansion or contraction.
iii. Thick enough to allow for any corrosion and/or erosion that will occur over
the life of the pipe.
iv. Thick enough to have threads cut in it, or to be flared or to be welded as the
case may be.
v. Not excessively thick as this represents waste of capital and resources.

Standards exist to assist in the determination of the required thickness such as

“AS4041 – 2006, Pressure Piping”, published 2006 by Standards Australia, and in
many applications are required to be adhered to by law.
Many manufacturers supply tables showing the safe working pressure of their
pipes as deduced from the Standards. However, the weight and corrosiveness of
the fluid used, the method of support used, and the other loads, if any, imposed on
the pipe, are beyond the control of the manufacturer, so the engineer must make
due allowance for this fact when using such tables provided by obliging
manufacturers. Such a table, Table 6.7.6 from Steel Pipe Design Manual, is
provided in Appendix 4A (Appendix 4A is located at end of this chapter) and refers
to the steel pipe shown in Table 7.2 the link to which is also given in Appendix 4A.
AS1074 referred to in these two tables is the Australian Standard for the quality,
size, etc. of that particular grade of pipe, and ASCB18 referred to in “Table 6.7.6”
was an early Australian Standard for determining the pressure rating, etc. of pipes.
Steel pipe to AS1074 with galvanized internal and external surfaces is commonly
used for compressed air pipes in factories in Australia. The galvanising provides
protection from corrosion caused by condensate. It can be seen from “Table 6.7.6”
that for threaded joints the allowed pressure becomes less as the diameter increases.
This will often mean that above a certain diameter it is necessary to have welded
rather than threaded joints. Threaded joints have the advantage that they are easy
to assemble and disassemble for small diameters but if poorly assembled are prone
to leak and waste air.

For the purposes of the assignment in this subject, pipe will be selected using these
two tables (Tables 7.2 and 6.7.6). In real life a more comprehensive analysis may be
necessary allowing for any extra loads that the pipes may experience. Such matters
are discussed in“AS4041 – 2006, Pressure Piping”

School of Engineering
Page 24 of 31
 Activity 4G
AS4041-2006: Recommended spacing of supports for steel pipe

Step1: Access the Australian Standard: AS4041- 2006 Pressure Piping

The Australian Standards can be accessed via the RMIT Library - SAI Global link
to the Australian Standards. 23

You will need to login with your student login and password.

Step 2: Refer to table 3.28.2 in Section 3 of AS4041-2006

Pipe must be supported at regular intervals so that: it is not overstressed through
bending; and doesn’t form sags where condensate can accumulate.

Please refer to Table 3.28.2 - Recommended spacing of supports for steel pipe
on page 141 of the standard as this provides recommendations on the frequency
of supports for steel piping.

Appendix 4A : Sample commercial steel pipe data

Table 7.2 Range of standard sizes tubeline pipe

Please refer to Table 7.2 of the Range of standard sizes tubeline pipe from page
76 of the Steel pipe design manual (PDF 71KB). 24

Table 6.7.6 AS1074 medium and heavy tube pressure and temperature ratings
- AS CB18 interpretation No. 1
Please refer to Table 6.7.6 AS1074 medium and heavy tube pressure and
temperature ratings - AS CB18 interpretation No. 1 from page 72 of the Steel
pipe design manual (PDF 101KB). 25

23 https://login.ezproxy.lib.rmit.edu.au/login?url=http://www.saiglobal.com/online/autologin.asp
24Tubemakers of Australia Limited, 1981, Table 7.2 Range of standard sizes tubeline pipe, Steel pipe design
manual, Tubemakers of Australia Limited Steel Pipe Division, Melbourne. p.76, viewed on 28th August 2013
<https://ap01-a.alma.exlibrisgroup.com/view/delivery/61RMIT_INST/12187201720001341.>

25Tubemakers of Australia Limited, 1981, Table 7.2 Range of standard sizes tubeline pipe, Steel pipe design
manual, Tubemakers of Australia Limited Steel Pipe Division, Melbourne. p.76, viewed on 28th August 2013
<https://ap01-a.alma.exlibrisgroup.com/view/delivery/61RMIT_INST/12187201740001341.>

School of Engineering
Page 25 of 31
Appendix 4B: Photos of common pipe and tube joining methods
Steel pipe is commonly joined by means of
threaded joints. Threads are cut into the
external surface of the pipe, whose walls must
be thick enough for this to be done. Such
threading is often done on site after the pipe
has been cut to length. The fitting into which
the pipe is being screwed has usually had
threads cut in its internal surface prior to
purchase. A sealing medium is applied on site
to the pipe threads before assembly. Teflon
tape is commonly used for this purpose.

Steel pipe is sometimes joined by means of

welding such as may be seen in the
illustration which shows a branch socket
welded to a pipe run. Welded joints are
usually leak tight however once done cannot
be easily disassembled.

Steel pipe is sometimes joined by means of a

pair of flanges. These flanges are either
screwed or welded to the steel pipe. A sheet
of gasket or “jointing” is prepared with holes
for the bolts and fluid cut into it. It is then
placed between the two flanges and the bolts
put in position and tightened up to an
appropriate tension. Flange joints provide a
very convenient point for disassembly and are
particularly useful for large diameter pipes
that are difficult to hold in alignment whilst
being rotated to screw together.

School of Engineering
Page 26 of 31
Copper tube is commonly joined together by
means of soldering. Both low and high
melting point solders are commonly used.
The tube is cut to length on site, deburred and
cleaned with an abrasive. Some solders
require a flux to be applied before assembly,
others are self fluxing. Once assembled, heat
is then applied to the joint typically with an
oxyacetylene torch. If nearby fittings may be
adversely affected by heat they will need
protection such as by being wrapped with a
wet rag. Soldered joints are usually leak tight
however once made cannot be easily
disassembled.

Copper tube is also commonly joined together

by means of flared joints. The tube must be
annealed prior to being flared. The flare nut
must be slid onto the tube before the tube is
flared. Flaring is then done with specially
shaped tools that hold the tube whilst a
conically shaped mandrel is driven into it
either propelled by a hammer or screw. Once
the flare has been formed the joint may be
assembled and the flare nut screwed up
firmly. Flared joints provide a convenient
point for disassembly.

Large copper tube is sometimes joined

together by means of pair of flanges.
In one type, as shown in the illustration to the
left, a specially flanged piece of copper is
soldered to the tube and then a much thicker
brass or steel flange is slid along the tube and
bolted to the joining fitting, holding the
flanged piece of copper firmly in place as it
does so.

In another type a heavy brass flange is brazed

onto the copper tube and then bolted to
adjoining fitting as shown in the illustration to
the left.
Flanged joints provide a convenient point for
disassembly.

School of Engineering
Page 27 of 31
 Small diameter copper tube is quite often
joined to fittings by means of a nut and
ferrule (aka “olive”) system. The tube is cut to
length on site then deburred. Care must be
taken not to scratch the outer surface of the
tube as this may provide a leakage path
between the tube and the ferrule The nut and
ferrule(s) are then slid onto the tube and the
joint assembled by pushing the tube against
an internal shoulder and tightening the nut.
As the nut is tightened it causes the ferrule to
bite into the surface of the tube providing a
seal. The two outer surfaces of the ferrule seal
against the nut and fitting respectively.

Flexible plastic tube and hose are also

commonly joined to fittings by means of a nut
and ferrule system. The fittings to which the
hose is joined incorporate a “barb” which the
hose is slid over up to the point where it butts
up against the shoulder. This barb provides
internal support to the tube as the nut is
tightened and the ferrule bites down onto the
tube and seals the tube inner surface against
the barb. An alternative to the nut and ferrule
is a corrosion inhibited steel band (“hose
clip”) that is tightened around the outside of
the tube using a screw system.

Plastic pipe is most commonly joined by

means of solvent cement which is applied to
the outer surface of the pipe and the inner
surface of the fitting before the two are
pushed together and held firmly until the
surfaces bond together. The surfaces must be
cleaned and primed before application of the
solvent cement.

Thick walled plastic pipe is sometimes joined

by means of threaded joints as shown in the
illustration and sometimes by plastic welding.

© RMIT University, 2013 (Marchiori G., Dixon, C.)

School of Engineering
Page 28 of 31
 Activity 4H Individual
Readings, Reflections and Calculations

Step 1: Refer to the Project Part B question sheets

Refer to the Project Part B question sheets. Read Question 1 entitled
DISTRIBUTION OF AIR TO TOOLS.

Step 2: Plan
Discuss with your partner the strategy for tackling Question 1

Step 3: Complete those parts of Question 1 that you are responsible for,
consulting with your partner where appropriate

Feedback
Feedback will be provided on your submitted project documentation by the
engineering lecturer/tutor responsible for marking it.

School of Engineering
Page 29 of 31
 Summary and Outcome Checklist

Mechanical, civil and chemical engineers frequently have to design piping or

ducting systems to distribute steam, water, air etc. Issues that the engineer
designing piping must address include:

(a) Deciding on pipe internal diameter. This must be:

(i) Large enough to pass the required flow without excessive frictional pressure
drop because large pressure drops mean wasted power and hence running
cost penalties.
(ii) Not so large that capital cost is being wasted on the pipe material, and the
contained fluid if that is expensive, as is the case for refrigerants.
(iii) Not so small that those velocities are high enough to create nuisance noise or
erosion.
(iv) Of such a size that any secondary fluids present are either dropped out of the
flow if that is desired, or carried along if that is desired.

(b) Deciding on the pipe material, and associated joining methods and
materials, which must:
(i) Be compatible with the fluid and not corroded too quickly.
(ii) Have reasonable strength at the operating temperature – not adversely
weakened by high temperature or embrittled by low temperature.
(iii) Be of reasonable first cost.
(iv) Be able to be fabricated at reasonable cost (e.g. threaded, flared or welded
joints can be made; bends can be made or fitted, etc.)

(c) Deciding on the pipe thickness which must be:

(i) Thick enough to safely carry the stresses caused by the fluid pressure
(ii) Thick enough to safely carry the extra stresses caused by other loads such
as: pipe weight and fluid weight, inertia loads where the fluid flow changes
direction, pressure pulses caused by the fluid flow being stopped suddenly,
and thrusts or pulls caused by thermal expansion or contraction.
(iii) Thick enough to allow for any corrosion and/or erosion that will occur
over the life of the pipe.
(iv) Thick enough to have threads cut in it, or to be flared or to be welded as
the case may be.
(v) Not excessively thick as this represents waste of capital and resources.

(d) Deciding on the appropriate location for the pipe which must:
School of Engineering
Page 30 of 31
 (i) Be convenient to the user.
(ii) Allow for future connections and extensions if likely.
(iii) Allow for drainage facilities if necessary.
(iv) Be accessible for maintenance purpose.
(v) Not take an excessively long route as this will both increase capital costs
due to the extra material used, and increase running costs due to the extra
frictional pressure drop.
(vi) Not be exposed to impact.
(vii)If possible, not be exposed to temperatures which could freeze its
contents. If the pipe does contain water or other fluids liable to freezing,
and has to be outside then it should be heavily insulated, fitted with
heating arrangements of some sort, buried or drained in freezing
conditions.
(viii)Protected from rain if this is liable to corrode it or damage insulation.

(e) Deciding on the method and frequency of supports, including allowance

for expansion/contraction.

(f) Deciding if the pipe should have a slope.

(g) Deciding if expansion joints are necessary for the pipe material, and
expansion vessels for the fluid.

(h) Decide if thermal insulation is required

(i) Deciding on what control and processing fittings should be incorporated

Tick the box for this statement if you agree with it:
 I can determine the layout and sizing of the compressed air piping system
including consideration of frictional pressure drop and condensate removal.

Assessment
This topic will be assessed as part of: Project Part B; and the mid-semester test
and/or the end of semester examination.

School of Engineering
Page 31 of 31