Technological University Dublin

ARROW@TU Dublin

Articles School of Mechanical and Design Engineering

2011-10-03

Technical Note: Friction Factor Diagrams for Pipe Flow

Jim McGovern
Technological University Dublin, [email protected]

Follow this and additional works at: https://arrow.tudublin.ie/engschmecart

Part of the Energy Systems Commons, Hydraulic Engineering Commons, and the Nanoscience and
Nanotechnology Commons

Recommended Citation
McGovern, J.: Technical Note: Friction Diagrams for Pipe Flow. Technological University Dublin, 2011.

This Working Paper is brought to you for free and open access by the School of Mechanical and Design
Engineering at ARROW@TU Dublin. It has been accepted for inclusion in Articles by an authorized administrator of
ARROW@TU Dublin. For more information, please contact [email protected], [email protected],
[email protected], [email protected].

Funder: National Digital Learning Resources and Technological University Dublin

 Technical Note: Friction Factor
Diagrams for Pipe Flow
Jim McGovern
Department of Mechanical Engineering
and Dublin Energy Lab
Dublin Institute of Technology, Bolton Street
Dublin 1, Ireland

Abstract
This technical note describes diagrams of friction factor for pipe flow
that have been prepared using, mainly, the equations that Lewis Moody
used to prepare his famous diagram in 1944. The preparation of the
new diagrams was prompted by the need for vector graphics versions
that could be used for teaching purposes and that could be distributed
freely to students and others under a Creative Commons Attribution-
Share-Alike license.
Using a structure very similar to that of Moody’s diagram, variants
with the Darcy friction factor and the Fanning friction factor have
been prepared. In addition, variants have been prepared that include
not only monotonic roughness curves, but also inflectional roughness
curves.
Keywords: friction factor, Moody diagram, vector graphic, Darcy
friction factor, Fanning friction factor, pipe flow, relative roughness,
Reynolds number, monotonic roughness, inflectional roughness.

Introduction
Since the mid 1940s, practicing engineers, engineering academics and stu-
dents of engineering have made use of a diagram of friction factor for pipe
flow of the type that was published by Lewis Moody in the Transactions
of A.S.M.E. in 1944 [1]. Textbooks in such areas as Fluid Mechanics, Hy-
draulics, Heat Transfer or Unit Operations commonly include re-drawn or
reproduced versions of Moody’s diagram. The diagram is semi-empirical,
based on some fundamental principles and the strong intuition of leading
researchers up to 1944. At every stage since then the diagram could be
regarded as a temporary solution, until sufficient further advance had been
made. Although considerable progress has been made, notably in computa-
tion, in the measurement of surface roughness and in the measurement and

1
Darcy Friction Factor for Pipe Flow Reynolds Number
103 1.4 2 3 4 5 6 104 1.4 2 3 4 5 6 105 1.4 2 3 4 5 6 106 1.4 2 3 4 5 6 107 1.4 2 3 4 5 6 108 1.4 2 3 4 5 6 109 ¥
0.08 0.08
0.06
0.055
0.05
0.07 0.07
0.045
0.04
0.06 0.035 0.06
0.03
0.025
Transition Zone
N

0.05 0.05
rV 2

0.02
2

0.018
0.045 0.016 0.045
= 4 tw ’ J

0.014
0.04 0.012 0.04
0.01
0.035 0.008 0.035
Complete Turbulence Zone
N

0.006
rV 2
2

0.03 0.005 0.03

Relative Roughness eD

0.028 0.004 0.028
= Dp’ J DL

0.026 0.003 0.026

Friction Factor
0.024 0.024
f f 0.002
0.022 ¥ 0.001 5 0.022
=
1.
01
N

0.02 0.001 0.02

2g
V2

0.018
an 0.000 8
0.018
d
Friction Factor fD = hf ’ J DL

0.000 6
Re
0.016 0.000 4 0.016
=
0.015 16 0.000 3 0.015
00
0.014 Critical He 0.014
D
0.000 2
Zone
0.013 L 0.013
0.012 Indicative Roughness Values 0.000 1 0.012
e @mmD
0.011
Smooth honed steel 0.00065
Sm 0.011
Drawn tubing: glass, brass, copper, lead, plastic 0.0015
o oth 0.000 05
0.01 0.01
Asphalted cast iron 0.12 Pip 0.000 03
0.009 Galvanized steel 0.15 e 0.000 02 0.009
Laminar Wood stave 0.18-0.91
Cast iron 0.26
0.008 Zone Concrete 0.3-3
0.000 01 0.008
fD = 64Re Heavy brush coat: asphalts, enamels, tars 0.45-0.6
0.000 005
General tuberculation 1-3 mm 0.6-1.9
0.007 0.007
Riveted steel 0.9-9 0.000 003
Severe tuberculation and incrustation 2.5-6.5 0.000 002
0.006 0.006
0.000 001

5. ´ 10-7
0.005 0.005
103 1.4 2 3 4 5 6 104 1.4 2 3 4 5 6 105 1.4 2 3 4 5 6 106 1.4 2 3 4 5 6 107 1.4 2 3 4 5 6 108 1.4 2 3 4 5 6 109 ¥

rVD VD
Jim McGovern, Dublin Institute of Technology Reynolds Number Re = m
= n
http://www.fun-engineering.net/pipefrictionfactor.html Version DM001 2011.10.02

Figure 1: Newly-prepared diagram of friction factor for pipe flow.

understanding of velocity distribution within boundary layers, the diagram

can still be regarded as ‘temporary,’ but continues to fulfill the function that
Moody attributed to it: ‘a simple means of estimating the friction factors.’
It still has relevance for a very broad range of situations, from flow in micro
tubes, e.g. reference [2], to flow in large pipelines or tunnels, e.g. references
[3, 4].
The equations that Moody used to prepare his diagram had been devel-
oped by others, as cited by Moody, and were supported by published data.
Figure 1 is a newly-prepared diagram of this type.
The friction factor f is a dimensionless term in the Darcy-Weisbach equa-
tion, Equation 1 or Equation 2. A concise history of the Darcy-Weisbach
equation has been written by Brown [5]. Two variants of the friction factor
are in common use: the Darcy friction factor fD , which Moody plotted, and
the Fanning friction factor fF , which equals one quarter of the Darcy factor,
as expressed in Equation 3. Versions of the diagram for both friction factors
have been prepared. !
L V2
hf = fD (1)
D 2g
!
L V2
hf = 4fF (2)
D 2g

2
 fF = fD /4 (3)
The fact that the friction factor diagram in the form set out by Moody
still endures is due to the use of dimensionless quantities and an effective
consensus to continue to use the constants and underlying equations that
were used by Moody. A consistent skeleton structure is provided. Moody
and his contemporaries were under no illusions about its accuracy. Moody
stated ‘it must be recognized that any high degree of accuracy in determining
f is not to be expected’ and ‘fairly reasonable estimates of friction loss can
be made.’
The diagram is necessarily a simplification and a rough approximation:
the conditions it describes (fully developed, isothermal, incompressible, dis-
sipative, pseudo-steady-state flow) are never quite attained in practice. Con-
ditions within pipelines are inherently non-uniform over the flow cross sec-
tion and over a given length. Fluids are more or less compressible, rather
than incompressible, and have finite thermal conductivity. The flowing fluid
within pipelines is not isothermal in the radial direction because of the very
dissipation that is quantified by the Darcy-Weisbach equation. Frictional
dissipation and finite thermal conductivity together give rise to differences
in temperature—and density and viscosity—between different parts of the
flow. Inevitably too, any length of pipe for which the fully-developed flow
requirement is approximated must be connected to entry and exit systems
within which this is not the case. When the flow is turbulent, vortices form
and collapse relentlessly over time. In the flow of liquids, vapour pressure
may play a role. Nevertheless, the diagram is a very useful design tool. For
computational purposes, it is easily represented as a data set of discrete
points, or by the equations that define it along with appropriate solution
algorithms.
The preparation of new diagrams by the present author was prompted by
the need for vector graphics versions of the friction factor diagram that could
be used for teaching purposes and could be distributed freely to students
and others under a Creative Commons Attribution-Share-Alike license. The
same four zones that were labeled by Moody (laminar, critical, transition
and complete turbulence) are shown. Colour has been employed, although
the diagram is also usable in grayscale form.

Preparation of the Diagrams

The diagrams were prepared using Wolfram Mathematica and exported (as
layers in PDF format) to CorelDraw for minor adjustments to the layout.
The nominal size of each of the diagrams is A3, in landscape format, and
printed versions at this size are very convenient to use. As most users will
not have a personal printer available for this size, care was taken to ensure

3
usability at A4 or Letter size. Absolute size is of little relevance for use on a
screen or tablet. The diagrams can be used conveniently on any device that
can display a PDF document.
The mathematical curves have not been plotted directly. Rather, straight
lines have been drawn between precisely calculated data points of the un-
derlying curves. The underlying equations, such as the Colebrook White
[6] equation, Equation 4, were solved or evaluated with a precision of about
fifteen significant digits. The plotted points are sufficiently close together
that visual smoothness is achieved and, as the diagram is a vector graphic,
the file size is small.
!
1 /D 2.51
p = −2 log10 + p (4)
fD 3.7 Re fD

The horizontal axis in the diagram is for the Reynolds number Re, Equa-
tion 5, which is dimensionless. The range has been extended at the upper
end to 109 (Moody’s diagram went to 108 ) and then to infinity!
ρ VD VD
Re = = (5)
µ ν
The vertical axis of the diagram is for the friction factor. The Darcy and
the Fanning friction factors are given by Equations 6 and 7 respectively.
! ! !
L V2 L ρV 2 ρV 2
fD = hf / = ∆p/ = 4τw / (6)
D 2g D 2 2

! ! !
L V2 L ρV 2 ρV 2
fF = hf / 4 = ∆p/ 4 = τw / (7)
D 2g D 2 2
Friction factors were in use prior to the publication of Moody’s diagram.
For example, in 1930 Fred Scobey [7] compiled ‘Weisbach friction coeffi-
cients,’ fD , for riveted and analogous pipes under actual operating condi-
tions. These friction factors were determined from measurements of flow
rate and pressure drop or head loss in situ.
For turbulent flow in smooth pipes, the friction factor is calculated from
the von Kármán, Prandtl, Nikuradse expression, Equation 8.
!
1 2.51
p = −2 log10 p (8)
fD Re fD

The friction factor in rough pipes at infinite Reynolds number is calculated

from von Kármán’s expression, Equation 9.
1 /D
 
p = −2 log10 (9)
fD 3.7

4
where /D is the relative roughness of the pipe,  being the surface absolute
roughness and D being the inside diameter of the pipe. Equations 8 and
9 are both included in the Colebrook White equation, Equation 4, which
achieves a blending of the two extremes to cover the full transitional and
complete turbulence zones of the diagram. Friction factor curves are shown
on the diagram for relative roughness values ranging from 1 × 10−7 to 0.06.
It can be noted that, in principle, the absolute roughness that is used is
an experimentally determined ‘equivalent sandgrain size’ parameter for the
particular pipe surface. It is referenced to tests carried out by J. Nikuradse
[8].

Transition / Complete Turbulence Boundary

In his diagram, Moody showed a dashed line to mark the boundary between
the transition zone and the complete turbulence zone. He plotted this by
making use of an expression that H. Rouse had used, Equation 10.
1 Re 
p = (10)
fD 200 D
However, within the written discussion at the end of his paper [1] Moody
mentioned that the boundary curve could be located ‘so that it would cor-
respond to some fixed percentage of excess in f over the f for complete
turbulence.’ This approach has been adopted by the present author: e.g. in
Figure 1, the boundary curve is defined by Equation 11.
f
= 1.01 (11)
f∞

As part of the discussion, R.J.S. Pigott, who had published a friction factor
diagram with basically the same structure as Moody’s in 1933 [9], proposed
Equation 12 for the boundary curve, while Moody responded with a sugges-
tion that Equation 13 would approximate to a boundary at 1% above the
friction factor for infinite Reynolds number. This has been verified by the
present author. Therefore, Equation 13 also represents the boundary curve
shown in Figure 1.
3500
= Re (12)
/D
1600
= Re (13)
/D
Hence, for a maximum positive correction factor of less than 1% in using
the rough pipes expression (Equation 9) for the friction factor, the complete
turbulence zone can be defined by the inequality labeled Equation 14.
1600
Re > (14)
/D

5
This can be generalized, with excellent precision, for an arbitrary positive
maximum correction factor, η, as the expression labeled Equation 15.

16 × 10−Log10 (η)
Re > (15)
/D

where 2.5 × 10−7 ≤ η ≤ 0.02, 104 ≤ Re ≤ 109 and /D > 0. For example,
setting η = 0.01 yields Equation 13. A practical application of this is to
avoid solving the Colebrook White equation where it is unnecessary. The
correction can be ignored, where it is insignificant, or applied (up to a max-
imum value of 0.02) according to Equations 16 and 17. This correction can
be applied and has a value less than 0.01 in the entire ‘Complete Turbulence
Zone’ of Figure 1.
f = (1 + η)f∞ (16)

η = 10−Log10 (Re D /0.16) (17)

Friction Factors for Smooth Pipes

Strong experimental data exist to support the friction factors in smooth
pipes over the range 11 ≤ Re ≤ 2.7 × 107 [8, 10, 2]. Laminar flow exists
up to a Reynolds number close to 3,000 and the flow is turbulent from
3,000 upwards. Furthermore, finely honed steel pipe has been found to be
hydraulically smooth at Reynolds numbers up to at least 2.7 × 107 [11],
which is far higher than the equivalent sand grain roughness would predict
according to the Colebrook White formula. Such pipe surfaces are said to
have an inflectional transition curve, which was also the case in the pipes
that were tested by Nikuradse [8] that had been artificially roughened with
sand grains.
Alternative equations to Equation 8 (which is also embedded in Equa-
tion 4) for the friction factor of smooth pipes have been proposed [10, 12, 13]
and there is some evidence to suggest that the smooth pipe friction factor
curve should be a little higher at high Reynolds numbers. However, corrob-
orating data are still scarce. Also, where friction factor has been measured
at very high Reynolds numbers the fluid has been a gas, which is not incom-
pressible. It is still possible that Equation 8 (based on the extrapolation of
test data for water) is good for a hypothetical incompressible fluid at very
high Reynolds numbers.

Determination of Absolute Roughness

The absolute roughness of a pipe could be measured by first determining
friction factor values from measurements of head loss and flow rate where
the flow is in the transition or complete turbulence zone and then solving

6
for  from Equation 4, as in Equation 18. Alternatively, the friction factor
data points could be superimposed on a Moody diagram in order to read the
relative roughness value and then calculate the roughness by multiplying by
the diameter. This was done, for example, in reference [3].
√
!
−1/2 fD 2.51
 = 3.7D 10 − p (18)
Re fD

However, as was well recognized by Colebrook and White [14], as well as

by Nikuradse and Moody, the nature of the surface and not just its ‘absolute
roughness’ influences the shape of the curves representing friction factor in
the transition region. Strictly, therefore, a mapping of particular pipes to
the relative roughness curves of the friction factor diagram should be based
on measurements made in the fully turbulent region.
In choosing to base the friction factor curves on the Colebrook White
equation, Moody choose to represent the characteristics of commercial pipes
(‘of the nature of cast iron, wrought iron or galvanized steel’—this phrase
is from Colebrook and White [14]) rather than artificially roughened pipes,
as had been tested by Nikuradse and by Colebrook and White. Recent
work at Princeton University [11] has shown that a honed surface can have
characteristics that are somewhat similar to those of Nikuradse’s surfaces
that were artificially roughened with sand.

Absolute and Relative Roughness Values

Moody [1] gave absolute roughness values and ranges for a selection of eight
pipe surfaces, based on published data. In seven of the eight cases the same
values and ranges had been shown on a friction factor diagram by Hunter
Rouse (1943), which was also included as part of the written discussion at
the end of Moody’s paper [1], on p. 681. Similar values had been included
in the 1938 paper by C.F. Colebrook [6].
In his 1933 friction factor diagram, Pigott [9] regarded pipes such as
those of drawn brass, lead or tin, or glass tubes, as being ‘as nearly dead
smooth as we can get them.’ Rouse’s diagram had shown ‘drawn’ tubing
(glass, brass, copper or lead) as ‘smooth,’ and Colebrook had also taken
these as smooth ‘at least for all ordinary velocities of flow,’ whereas Moody
gave a finite value, equivalent to 0.0015 mm, based on two cited references.
A somewhat similar situation has arisen in relation to the smoothness of the
original pipe of the ‘SuperPipe’ experiment at Princeton University [15].
This pipe was found to be smooth up to a very high Reynolds number.
Some indicative relative roughness values from the literature are shown
in Table 1. Values such as these are commonly listed in textbooks and
handbooks, e.g. [16]. Surface finishes can vary considerably and it is also

7
important to take account of condition, corrosion or fowling. The summary
table for inclusion on the printed diagram is shown as Table 2.
Relative roughness values depend on the pipe diameter and, for rough
pipes, there could be some ambiguity about which ‘diameter’ to use—a
volume based average would seem the most appropriate, as in Equation 19.
‘Roughness’ would thus extend inside and outside the ‘diameter.’
s
4V
D= (19)
πL

Depending on the circumstances, different methodologies for measuring the

diameter may be used, but the approach used by Nikuradse [8] was exem-
plary:

‘The diameters of the pipe were determined from the weight of

the water which could be contained in the pipe with closed ends
and from the length of the pipe.’

Pipes with Inflectional Roughness

The roughness curves derived from the Colebrook White equation are said to
be monotonic, i.e. the friction factor decreases continuously with increasing
Reynolds number. In the tests carried out by Nikuradse on pipes that were
artificially roughened with grains of sand the curves were inflectional in
nature, i.e. the friction factor decreases to a minimum value with increasing
Reynolds number and then rises again to reach a constant value for complete
turbulence. Some modern pipes, e.g. honed steel pipe, demonstrate similar
inflectional friction factor curves. Afzal [13] has provided an extension of
the Colebrook White equation that is capable of representing Nikuradse’s
data and data from modern pipes that have inflectional friction factor curves
with good accuracy, Equation 20. This is identical to the Colebrook White
equation when the dimensionless roughness parameter j has a value of zero.
Afzal found that a value of 11 for j provided a good fit to the data of
Sletfjerding and Gudmundsson [17], Shockling [18, 11], and Nikuradse [8].
The version of Equation 20 given in [13] appears to omit a factor of two
in the exponential term. The version given here has been checked by the
present author against the experimental data to confirm good agreement.
!
1 /D 2.83 2.51
 
p = −2 log10 exp −j √ + p (20)
fD 3.7 Re(/D) f Re fD

Figure 2 is a is a variant of Figure 1 that includes inflectional roughness

curves in accordance with Equation 20, where j has a value of 11.

8
Darcy Friction Factor for Pipe Flow Reynolds Number
103 1.4 2 3 4 5 6 104 1.4 2 3 4 5 6 105 1.4 2 3 4 5 6 106 1.4 2 3 4 5 6 107 1.4 2 3 4 5 6 108 1.4 2 3 4 5 6 109 ¥
0.08 0.08
0.06
0.055
0.05
0.07 0.07
0.045
0.04
0.06 0.035 0.06
0.03
0.025
Transition Zone
N

0.05 0.05
rV 2

0.02
2

0.018
0.045 0.016 0.045
= 4 tw ’ J

0.014
0.04 0.012 0.04
0.01
0.035 0.008 0.035
Complete Turbulence Zone
N

0.006
rV 2
2

0.03 0.005 0.03

Relative Roughness eD

0.028 0.004 0.028
= Dp’ J DL

0.026 0.003 0.026

Friction Factor
0.024 0.024
f f 0.002
0.022 ¥ 0.001 5 0.022
=
1.
01
N

0.02 0.02
Sample 0.001
2g
V2

0.018
inflectional an 0.000 8
0.018
d
Friction Factor fD = hf ’ J DL

roughness 0.000 6
curves Re
0.016 0.000 4 0.016
=
0.015 16 0.000 3 0.015
00
0.014 Critical He 0.014
D
0.000 2
Zone
0.013 L 0.013
0.012 Indicative Roughness Values 0.000 1 0.012
e @mmD
0.011
Smooth honed steel 0.00065
Sm 0.011
Drawn tubing: glass, brass, copper, lead, plastic 0.0015
o oth 0.000 05
0.01 0.01
Asphalted cast iron 0.12 Pip 0.000 03
0.009 Galvanized steel 0.15 e 0.000 02 0.009
Laminar Wood stave 0.18-0.91
Cast iron 0.26
0.008 Zone Concrete 0.3-3
0.000 01 0.008
fD = 64Re Heavy brush coat: asphalts, enamels, tars 0.45-0.6
0.000 005
General tuberculation 1-3 mm 0.6-1.9
0.007 0.007
Riveted steel 0.9-9 0.000 003
Severe tuberculation and incrustation 2.5-6.5 0.000 002
0.006 0.006
0.000 001

5. ´ 10-7
0.005 0.005
103 1.4 2 3 4 5 6 104 1.4 2 3 4 5 6 105 1.4 2 3 4 5 6 106 1.4 2 3 4 5 6 107 1.4 2 3 4 5 6 108 1.4 2 3 4 5 6 109 ¥

rVD VD
Jim McGovern, Dublin Institute of Technology Reynolds Number Re = m
= n
http://www.fun-engineering.net/pipefrictionfactor.html Version DMI001 2011.10.02

Figure 2: Diagram of friction factor for pipe flow, including sample inflec-
tional roughness curves.

Absolute Roughness of Modern Pipe Surfaces

Precise information for friction factor versus Reynolds number is available
for two specific pipe types from publications relating to the Princeton Su-
perPipe tests [15, 19, 11]. Of these, the ‘rough’ pipe displays an inflectional
characteristic in the transition zone, which is similar to the inflectional char-
acteristic of the sand-grain-roughened pipes that were tested by Nikuradse
[8]. The data for the ‘smooth’ pipe all lie along the left-hand edge of the
transitional zone and no inflectional friction factor characteristic is apparent
up to the highest Reynolds number of 3.5 × 107 .
Matching Equation 20 to the Princeton SuperPipe ‘rough pipe’ data sug-
gests that the inner surface had an equivalent sand grain roughness of about
10 micron or 0.01 mm—the measured rms was 2.5 micron or 0.0025 mm.
On the same basis, the sand grain roughness of the Princeton SuperPipe
‘smooth pipe’ would have had to be less than about 0.65 micron, or 0.00065
mm—the measured root mean square roughness, rms , was 0.15 micron.
Sletfjerding and Gudmundsson [17] provided test results (using dry nat-
ural gas) for pipes artificially roughened with glass beads and epoxy, which
had inflectional friction factor curves, as well as for a honed steel pipe and
an epoxy coated smooth pipe. An equivalent sand grain roughness of 21 mi-
cron was determined for the honed steel pipe, which had a measured rms

9
of 3.66 micron. An equivalent sand grain roughness was not determined for
the smooth coated pipe, which had a measured rms of 1.41 micron.
There is still a scarcity of data on the absolute sand-grain-equivalent
roughness of commercial pipes. The values provided by Farshad et al. [20,
21, 22] are illustrative. An examination of the data available in [21] indicates
that flow measurements were made at a Reynolds number of about 6.5 × 105
and in all cases were within the transitional zone. Also, in [22] the authors
seem to have misinterpreted Moody’s chart (his Figure 2, p. 678) that related
relative roughness to pipe diameter for various absolute roughness values—
Moody’s log-log chart of /D versus D was simply based on the inverse
proportionality between relative roughness and pipe diameter.

Using the Diagram

When the diagram is printed-out, a straight edge can be used to assist in
reading the value of the friction factor on the scales that are provided on the
left and right edges. When using the diagram on a screen or tablet, users
may find it helpful to use the rulers and movable guidelines that are usually
available within a PDF reader program.

Conclusions
The diagrams have been prepared by the author for students of engineering
and practicing engineers. He would welcome any suggestions for further
adjustments.

Acknowledgement
This work was supported in part by a National Digital Learning Resources
and Dublin Institute of Technology Award within the 2010–2011 academic
year.

Nomenclature
∆p pressure drop due to fluid friction

 absolute surface roughness

(equivalent sand grain roughness)

rms root mean square surface roughness

η correction factor

V volume

10
Re Reynolds number

µ absolute viscosity

ν kinematic viscosity

ρ density

τw shear stress at the pipe wall

A cross-sectional area of pipe

D pipe diameter

f friction factor

f∞ friction factor at infinite Reynolds number

fD Darcy friction factor

fF Fanning friction factor

g acceleration due to gravity

hf head loss due to fluid friction

j dimensionless roughness parameter

L pipe length

Q volume flow rate

V mean flow velocity = Q/A

ρV 2 /2 dynamic pressure corresponding to mean flow velocity

V 2 /2g velocity head corresponding to mean flow velocity

PDF portable document format

11
 Table 1: Indicative absolute roughness values
Ref. /[mm]
Finely honed smooth steel
(rms = 0.00015 mm, inflectional) 11 (2007) 0.00065
Drawn tubing (e.g. glass, brass,
copper, lead, plastic) 1 (1944) 0.0015
Butt-welded steel: new smooth pipe
or with centrifugally applied enamels 3 (1977) 0.009 - 0.06
Honed steel
(rms = 0.0025 mm, inflectional) 11 (2007) 0.01
(rms = 0.0037 mm, inflectional) 17 (2003) 0.021
Butt-welded steel: centrifugally
applied concrete linings 3 (1977) 0.045 - 0.15
Wrought iron, steel 1 (1944) 0.046
Butt-welded steel:
hot asphalt dipped 3 (1977) 0.06 - 0.15
Asphalted cast iron 1 (1944) 0.12
Galvanized iron 1 (1944) 0.15
Butt-welded steel: light rust 3 (1977) 0.15 - 0.35
Wood stave 1 (1944) 0.18 - 0.91
Cast iron 1 (1944) 0.26
Concrete 1 (1944) 0.3 - 3
Butt-welded steel: heavy brush coat,
asphalts, enamels and tars 3 (1977) 0.45 - 0.6
Riveted steel 1 (1944) 0.9 - 9
Butt-welded steel:
general tuberculation 1-3 mm 3 (1977) 1 - 1.85
severe tuberculation
and incrustation 3 (1977) 2.5 - 6.5

12
 Table 2: Summary table of indicative absolute roughness values
/[mm]
Smooth honed steel 0.00065
Drawn tubing: glass, brass, copper, lead, plastic 0.0015
Asphalted cast iron 0.12
Galvanized steel 0.15
Wood stave 0.18 - 0.91
Cast iron 0.26
Concrete 0.3 - 3
Heavy brush coat: asphalts, enamels, tars 0.45 - 0.6
General tuberculation 1-3 mm 0.6 - 1.9
Riveted steel 0.9 - 9
Severe tuberculation and incrustation 2.5 - 6.5

References
[1] Lewis F. Moody. Friction factors for pipe flow. Transactions of the
A.S.M.E., pages 671–684, November 1944.

[2] Marco Lorenzini, Gian Luca Morini, and Sandro Salvigni. Lami-
nar, transitional and turbulent friction factors for gas flows in smooth
and rough microtubes. International Journal of Thermal Sciences,
49(2):248–255, 2010.

[3] J.N. Bradley, L.R. Thompson, C.W. Thomas, R.B. Dex-

ter, J.C. Schuster, A.J. Peterka, and H.M. Martin. Fric-
tion factors for large conduits flowing full, Division of Re-
search, Engineering Monograph No. 7. Technical Report
EM-7, 1977. Available at http://www.usbr.gov/pmts/
hydraulics lab/pubs/EM/EM07.pdf, accessed 4 August 2011.

[4] C. Montecinos and K. Wallace Jr. Equivalent roughness for pressure

drop calculations in mine ventilation. In Hardcastle and McKinnon,
editors, 13th United States/North American Mine Ventilation Sympo-
sium, Sudbury, Ontario, pages 225–230. MIRARCO, June 2010.

[5] Glenn O. Brown. The history of the Darcy-Weisbach equation for pipe
flow resistance. In A. Fredrich and J. Rogers, editors, Proceedings of
the 150th Anniversary Conference of ASCE, Washington D.C., pages
34–43. American Society of Civil Engineers, November 3-6 2002. Avail-
able at http://bioen.okstate.edu/Darcy/DarcyWeisbach/
HistoryoftheDarcyWeisbachEq.pdf, accessed 2 August 2011.

[6] C.F. Colebrook. Turbulent flow in pipes, with particular reference to

13
 the transition region between the smooth and rough pipe laws. Journal
of the Institution of Civil Engineers, 11:133–156, 1938-1939.

[7] Fred C. Scobey. The flow of water in riveted steel and analogous pipes.
United States Department of Agriculture, January 1930. Available at
http://naldr.nal.usda.gov/, accessed 8 August 2011.

[8] J. Nikuradse. Laws of flow in rough pipes (English translation

of Strömungsgesetze in rauhen Rohren, VDI-Forschungsheft 361,
Ausgabe B, Band 4, pp. 1–22, July/August 1933), pp. 63, NACA Tech-
nical Memo 1292, Washington. November 1950. Translation available at
http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov
/19930093938 1993093938.pdf, accessed 16 August 2011.

[9] R.J.S. Pigott. The flow of fluids in closed conduits. Mechanical Engi-
neering, 55:497–501,515, August 1933.

[10] B.J. McKeon, C.J. Swanson, M.V Zagarola, R.J. Donnelly, and A.J.
Smits. Friction factors for smooth pipe flow. Journal of Fluid Mechan-
ics, 511:41–44, 2004.

[11] J.J. Allen, M.A. Shockling, G.J. Kunkel, and A.J. Smits.
Turbulent flow in smooth and rough pipes. Philosophical
Transactions. Series A, Mathematical, physical, and engineer-
ing sciences, 365(1852):699–714, March 2007. Available at
http://rsta.royalsocietypublishing.org/cgi/content/
abstract/365/1852/699, accessed 22 August 2011.

[12] B.J. McKeon, M.V. Zagarola, and A.J. Smits. A new friction factor
relationship for fully developed pipe flow. Journal of Fluid Mechanics,
538:429–443, 2005.

[13] N. Afzal. Friction factor directly from transitional roughness in a tur-

bulent pipe flow. Journal of Fluids Engineering, 129:1255–1267, 2007.

[14] C.F. Colebrook and C.M. White. Experiments with fluid friction in
roughened pipes. In Proceedings of the Royal Society of London A,
Volume 161, pages 367–381. The Royal Society, February 1937.

[15] Princeton University. Princeton gas dynamics lab facilities:

The SuperPipe facility, 2011. http://www.princeton.edu
/[tilde]gasdyn/Facilities/Facilities.html, accessed 21
August 2011.

[16] Chartered Institution of Building Services Engineers. Guide C: Refer-

ence Data. CIBSE, London, 2007.

14
[17] Elling Sletfjerding and Jon Steinar Gudmundsson. Friction factor di-
rectly from roughness measurements. Journal of Energy Resources
Technology, Transactions of the ASME, 125(2):126–130, 2003.

[18] J.J. Allen, M.A. Shockling, and A.J. Smits. Evaluation of a universal
transitional resistance diagram for pipes with honed surfaces. Physics
of Fluids, 17(12):121702, December 2005.

[19] M.A. Shockling, J.J. Allen, and A.J. Smits. Roughness effects in tur-
bulent pipe flow. Journal of Fluid Mechanics, 564(1):267–285, 2006.

[20] Fred Farshad, Herman Rieke, and James Garber. New developments in
surface roughness measurements, characterization, and modeling fluid
flow in pipe. Journal of Petroleum Science and Engineering, 29:139–
150, February 2001.

[21] Fred F. Farshad, H.H. Rieke, and Charlie Mauldin. Flow test validation
of direct measurement methods used to determine surface roughness in
pipes (OCTG). In SPE Western Regional/AAPG Pacific Section Joint
Meeting, pages 529–540, Anchorage, AK, United States, May 20–22
2002.

[22] F.F. Farshad and H.H. Rieke. Surface-roughness design values for mod-
ern pipes. SPE Drilling & Completion, pages 212–215, September 2006.

Version 1, October 3, 2011

15