Pipelines 2010: Climbing New Peaks to Infrastructure Reliability—Renew, Rehab, and Reinvest 381

© 2010 ASCE

Mechanical Property Changes in Steel during the Pipe Making Process

Brent Keil1

Abstract
Welded Steel Pipe (WSP) is arguably the most widely utilized pipe material for the
transmission of water throughout the United States. Spiral WSP is manufactured by
placing steel coils, supplied by steel mills to the pipe manufacturer, into a pipe mill
that levels the coils and helically rolls and welds the steel sheet into pipe. During the
pipe manufacturing process, the steel undergoes various permanent changes in shape.
These changes require the steel to yield, which may result in alterations to the
mechanical properties of the original steel coil. The pipe ends may also be expanded
past yield point or swaged to shape a bell, while spigot ends may be rolled to
accommodate gaskets. All of this processing can have varying effects on the steel’s
mechanical properties. While various factors of safety incorporated into the design of
steel pipe systems amply address any potential mechanical property changes that may
take place during fabrication of the pipe, the topic has never been fully understood by
the design engineering community at large. In order to address this issue, testing was
done to measure the changes in steel mechanical properties during the pipe
manufacturing process. This paper outlines the testing and the subsequent variations
found in the mechanical properties of the steel used through the manufacturing
process. Based on the testing, the paper will provide conclusions and design
considerations for pipe with both restrained (welded) and non-restrained joints.

Introduction
It is well understood that cold working steel will change its mechanical properties.
The amount of change that can occur is dependant on the amount of work done to the
steel. The actual amount of change that can be expected when forming steel pipe has
been explored and studied in the oil and gas steel pipe market, but not much work on
the subject has been completed in the water pipe market. Although the testing for the
Pipeline Research Council International by Battelle Memorial Institute (Maxey 1988)
as well as research done by Shoemaker (1984) showed no impact to the function or
design of the pipe, the research that had been completed on pipe was manufactured by
the U-ing and O-ing processes (See Figure 1) to oil and gas pipe standards.

Figure 1 U-ing and O-ing Pipe Making Process

1
P.E., Corporate Chief Engineer, Northwest Pipe Co., 5721 SE Columbia Way, Suite
200, Vancouver, WA 98661, Tel: (360) 397-6250, Email: [email protected]

1
Pipelines 2010: Climbing New Peaks to Infrastructure Reliability—Renew, Rehab, and Reinvest 382
© 2010 ASCE

There has been interest and misunderstanding in the steel water pipe marketplace
regarding the actual amount of change that can occur in the steel in a typical
manufacturing process. There are differences in steel water pipe manufacturing when
compared to oil and gas pipe in that it is typically made by helically shaping steel
from coil. Additional steps are often used to form a bell and spigot in the steel pipe
for water applications that oil and gas pipe would not typically see. Because of the
different characteristic of the steel typically used in oil and gas pipe, the different
forming process used and the added steps typical in the water industry, questions have
been raised about the potential mechanical property changes and its impact on the
steel pipe’s performance over time.

To address questions regarding the effects of cold forming during the typical steel
water pipe manufacturing process, testing was conducted to determine mechanical
changes found in the steel. Tests were completed on three different pipe sections. One
pipe sample was from 84” nominal diameter, 0.625” wall thickness with a weld bell
joint (325 psi design operating pressure), and the other two were 30” nominal
diameter rolled groove gasket joint pipes one 0.149” (165 psi design operating
pressure) and one 0.215” (300 psi design operating pressure) thickness. The three
pipes were manufactured to AWWA C200 standards using the spiral forming method.
Testing conducted on all of the samples were Charpy impact, yield strength, ultimate
tensile strength, elongation, hardness, and grain structure.

Pipe Manufacturing Process

The most commonly used pipe manufacturing method for steel water pipe is spiral
welding coil steel in a continuous process (See Figure 2). The process starts with steel
coils placed in the pipe mill and the material unrolled off the coil and leveled. The
edges of the coil are trimmed or machined for welding and the steel is fed through
forming rolls at a set helix angle. The feed angle and the width of the coil determine
the diameter of the pipe. The pipe is formed and welded in a continuous operation and
typically joined with submerged arc welding, with a weld on the inside of the pipe
and one on the outside. The pipe sections are then cut at the desired length.

Figure 2 Spiral Pipe Manufacturing

Once the pipe barrel is formed, the ends are formed to accommodate connections to
other pipe sections during installation. The bell end typically is formed by

2
Pipelines 2010: Climbing New Peaks to Infrastructure Reliability—Renew, Rehab, and Reinvest 383
© 2010 ASCE

hydraulically expanding the pipe end with a segmented die. Another common method
entails the bell being formed by pushing the pipe end over a die, taking the steel past
yield point, which is sometimes called swaging. With swaging, the desired bell shape
and size are determined by the die. For a weld bell joint, the pipe end is typically
expanded to increase the diameter by twice the steel thickness, to allow for
engagement of the mating spigot (see Figure 3). Bells formed for gasket joints are
typically swaged or expanded further than weld bell joint to accommodate the gasket
retaining system without reducing the inside diameter of the pipe at the spigot end
(see Figure 4).

Figure 3 Typical Weld Bell Joint

Figure 4 Typical Rolled Groove Gasket Joint

If the pipe is a rolled groove gasketed pipe, the next step is to roll the spigot shape
that holds the gasket in place. This is done by rolling the spigot end of pipe through
shaped dies.

It can be readily seen that all of these processes cold work the steel to one degree or
another.

Testing Procedures
Samples were gathered from 0.625”, 0.207” and 0.149” thick material. For the weld
bell pipe, samples were taken from the coil before fabrication, from the pipe body
after forming, and from the bell after expansion. The bell was expanded using a
hydraulic segmented expander.

The rolled groove gasketed pipe had samples taken from the pipe body, the expanded
bell and from the formed gasket section on the spigot. The expanded bells on the
gasketed pipe were formed by the swedge method. The samples were cut from the
specific areas and each sample tested for the following:
1. Charpy v-notch testing per ASTM A370-09a Standard Test Methods and
Definitions for Mechanical Testing of Steel Products.
2. Rockwell hardness “B” testing per ASTM E18-08b Standard Test Methods for
Rockwell Hardness of Metallic Materials
3. Grain size per ASTM E112-96 Standard Test Methods for Determining
Average Grain Size
4. Yield point per ASTM A370-09a Standard Test Methods and Definitions for
Mechanical Testing of Steel Products

3
Pipelines 2010: Climbing New Peaks to Infrastructure Reliability—Renew, Rehab, and Reinvest 384
© 2010 ASCE

5. Ultimate tensile strength per ASTM A370-09a Standard Test Methods and
Definitions for Mechanical Testing of Steel Products
6. Elongation per ASTM A370-09a Standard Test Methods and Definitions for
Mechanical Testing of Steel Products

Yield point was determined by using the 0.2% offset method in accordance with
ASTM A370. Elongation was determined by using the 2 inch gage length for the
0.149” and 0.207” thickness materials. The 0.625” material elongation was tested
using the bar method, giving equivalent results to a 2 inch gage, as prescribed in
ASTM A370.

Rockwell hardness tests were reported as method “B” as described in ASTM E18 for
all samples.

Grain size was measured using 200X magnification using ASTM A112.

Charpy test temperatures for all samples was 30 Deg F. Full size samples were tested
for the 0.625” thickness material and subsize samples were required for the 0.207”
and 0.149” thickness materials as detailed in ASTM A370.

Tested Materials
The three steel chemistries are representative of some of the most common grades of
steel used in steel water pipe in the United States. The materials tested were as
follows:
1. 0.625” thickness coil manufactured in accordance with ASTM A 1018, Grade
45, Class 1, high strength low alloy steel (HSLAS).
2. 0.207” thickness coil manufactured in accordance with ASTM 1011, Grade
45, structural steel (SS).
3. 0.149” thickness coil manufactured in accordance with ASTM 1011, Grade
36, Type 1, structural steel (SS).

The ASTM minimum required yield, tensile and elongation for the steel grades tested
can be found in table 1.

Grade Thickness Yield Tensile Elongation 2"

1018 Gr 45 Cl1 0.625 45,000 60,000 22
1011 Gr 45 SS 0.207 45,000 60,000 19
1011 Gr 36 Type 1 0.149 36,000 53,000 22

Table 1 ASTM Minimum Requirements for Tested Grades of Steel

All materials were ordered in accordance with the listed ASTM standard as well as
the additional requirements of AWWA C200. Additionally, AWWA C200 requires
steel to be fully killed, and manufactured to a fine grain practice. ASTM defines fine
grained to have a grain size number of 5 or greater as determined by ASTM E112.

4
Pipelines 2010: Climbing New Peaks to Infrastructure Reliability—Renew, Rehab, and Reinvest 385
© 2010 ASCE

The pipe, bells, and spigots were formed and fabricated in accordance with AWWA
C200. AWWA C200 further requires the ordered yield, tensile, and elongation
properties of the steel are minimum values. As such, the actual values seen from steel
mills tend to be considerably higher than those ordered because the steel mills must
certify the material does not drop below those minimum values. Because steel pipe
designed to the AWWA standards is based on the minimum ordered yield strength of
the steel, the additional level of mechanical properties provided from the steel mill
gives safety factors that are not utilized during pipe design, giving steel pipe an
additional level of conservatism in its application.

The level of cold work applied to the steel for typical pipe can be expected to increase
depending on the joint configuration. The lowest expected level of cold working is
found in straight pipe with no joint preparation completed. Next is the expanded weld
bell end followed by the swaged or expanded gasket bell end and the rolled groove
getting the most cold work of the joints tested. The gasket bell typically receives
more cold work than a weld bell because it must be expanded further than a weld bell
to accommodate the rolled shape of the spigot.

Sample Preparation
Samples were flame cut from the coil, pipe barrel and pipe bell and spigot. Each
sample was identified for traceability by a sample number, coil number and location
from where they were removed. Figure 5 shows a typical sample with it’s
identification before removal from a pipe section.

Figure 5 Typical Sample Layout and Marking (Rolled Groove Spigot)

5
Pipelines 2010: Climbing New Peaks to Infrastructure Reliability—Renew, Rehab, and Reinvest 386
© 2010 ASCE

Figure 6 shows the removal of a sample from the pipe. Once the samples were
removed from the pipe they were submitted to an independent testing lab for
evaluation. All testing was completed by IMR KHA of Portland Oregon, an ISO/IEC
17025 certified lab. Samples were prepared in accordance with the applicable ASTM
standards for the tests listed.

Figure 6 Typical Sample Removals

Samples in the pipe body were taken both in the steel rolling direction as well as in
the pipe hoop direction in the 0.149” and 0.207” samples. Samples from the pipe bells
were taken in the flat of the bell (the area of engagement with the mating pipe).
Rolled groove samples were taken to include the rolled shape of the spigot and the
test samples were all prepared in a manner to include the worked steel in the results.

It should be noted, the samples from the rolled groove area required additional
preparation beyond that of the other samples, Figure 7 shows the rolled groove shape.
It was necessary to flatten the rolled shape before testing. For the yield / tensile
samples a section had to be welded on to the end of the sample to get the formed end
inside the tested area. The step of flattening added an additional step of cold working
that would not be seen in actual application.

Figure 7 Typical Rolled Groove Shape

6
Pipelines 2010: Climbing New Peaks to Infrastructure Reliability—Renew, Rehab, and Reinvest 387
© 2010 ASCE

Testing Results
Tables 2, 3 and 4 give the test results from each of the samples. The data was
tabulated, each test location values were averaged and all the values from each
thickness of material were averaged and a standard deviation was calculated.

For the Charpy results (CVN) the actual result is listed (CVN AVG) and the sample
size is listed. Standard full size Charpy samples are 10 mm by 10 mm, if the material
being sampled is below that thickness, subsize samples are tested. The 0.149”
required 10 mm by 2.5 mm or ¼ size samples and 0.207” required 10 mm by 5 mm or
½ size samples. Although a direct meaningful correlation of subsize to full size values
is difficult to make, an approximation can be made based on ASTM A370. The
approximation is made by simply dividing the subsize Charpy value by its
corresponding size. For example a ½ size sample that achieved 15 ft-lbs would have
a corrected value of 30 ft-lbs. The approximations for the subsize samples tested are
listed as CVN Corrected For Subsize in Tables 3 and 4.

Figures 8 and 9 show micrograph photos of 0.149” samples magnified 200X. The
samples were taken from the pipe body and the rolled spigot. It can be seen that there
is no discernable change in the grain structure of the steel.

Figure 8 0.149” Pipe Body Sample Figure 9 0.149” Rolled Groove Sample

Conclusions
The test results after cold forming of pipe barrel, expanded bells and rolled spigots
show that the as manufactured steel properties demonstrate no discernable reduction
in steel properties and that the as tested steel properties are far greater than those
required by AWWA standards.

The testing data does demonstrate some minor variability in the results, particularly in
the yield strength of the steel. It is unclear if the variability in the testing results is due
to the natural properties of the steel as supplied from the steel mill or if there was
some change induced from the pipe making process. It is well known that cold
working steel will increase the tensile-yield strength, and therefore some of the minor
changes seen are likely attributed to the cold working.

7
Pipelines 2010: Climbing New Peaks to Infrastructure Reliability—Renew, Rehab, and Reinvest 388
© 2010 ASCE

The changes in yield strength are insignificant in typical water pipe applications, first
because the amount of possible change seen in the testing would be considered
normal variance in properties as received from the steel mills. Second, the change in
yield strength from cold working will be an increase, not a decrease, adding to the
design safety factor of the pipeline. Ultimately the amount of variation seen in the
tests is negligible in pipe applications.

There were no discernable changes in the Charpy, Rockwell hardness, ultimate

tensile, elongation, or grain structure properties tested. The results were within the
expected accuracy of the tests completed. It should be noted that there is some
amount of variance expected simply from the test process itself. For example, in
Charpy testing, variability of a few ft-lbs is considered normal due to the changes that
can come from the testing process. Slight variations in the radius preparation of the
notch placed in the steel, the thickness variation of the sample, variations in
temperature of the sample at time of impact, the placement of the sample in the
testing apparatus. All of these variables have allowable tolerances in the ASTM
standard and they can have mirror effects on the test values. Variations of a few ft-lbs
are not uncommon in Charpy results. It is one of the key reasons that ASTM methods
require that three samples are averaged to create one net Charpy test resultant. This
type of variation can be easily misinterpreted when the Charpy values are particularly
low. A 2 or 3 ft-lb variance in samples breaking at 10 or 11 ft-lbs force looks much
more significant than the same variance at 50 or 100 ft-lbs.

It was thought the rolled groove spigot would show the greatest variances since it not
only underwent the most cold working in forming the shape, but it was worked again
to flatten the sample for testing. The testing results did not reflect this assumption.
The test results show that the steel tested is capable of being rolled and flattened
without significant changes to the mechanical properties of the steel.

Based on the results of the testing, mechanical properties of steel during typical pipe
manufacturing processes used for pipe manufactured to AWWA C200 do not change
in any significant way. Further, design safety factors applied to steel water pipe are
not reduced by the pipe manufacturing process. Steels used in the testing show that
they are capable of handling the cold work applied during the manufacturing process
of AWWA C200 steel water pipe without degradation of their mechanical properties
from pipe making or joint formation. The testing confirms there is no need to limit
hoop stress or axial load design for unrestrained rolled groove or restrained weld bell
joints.

8
Pipelines 2010: Climbing New Peaks to Infrastructure Reliability—Renew, Rehab, and Reinvest 389
© 2010 ASCE

Table 2 Test Results for 0.625” thickness 84” Diameter Weld Bell Pipe
CVN
Yield Tensile CVN CVN
Thickness Diameter Sample Sub- Corrected Rockwell Grain
psi psi T-Y Y/T Elongation AVG Size
in. in. Location size For Sub- B Avg Size
(Y) (T) Ft-lbs mm
size
0.625 84 Coil 52200 65700 13500 0.79 32.0 262.51 10 N NA 75.0 8.0
0.625 84 Coil 53200 65900 12700 0.81 32.5 262.51 10 N NA 77.0 10.0
0.625 84 Coil 56200 66500 10300 0.85 31.5 262.51 10 N NA 77.0 9.5
Average 53867 66033 12167 0.82 32.0 262.5 76.3 9.2

0.625 84 Pipe Body 54300 66700 12400 0.81 34.0 262.51 10 N NA 77.5 9.5
0.625 84 Pipe Body 54000 67000 13000 0.81 32.0 262.51 10 N NA 77.5 9.5
0.625 84 Pipe Body 53500 67200 13700 0.80 31.0 262.51 10 N NA 77.5 9.5
Average 53933 66967 13033 0.81 32.3 262.5 77.5 9.5

0.625 84 Weld Bell 53300 67000 13700 0.80 30.5 262.51 10 N NA 77.5 9.0
0.625 84 Weld Bell 54500 67200 12700 0.81 32.0 262.51 10 N NA 79.0 9.5
0.625 84 Weld Bell 53400 66700 13300 0.80 31.0 262.51 10 N NA 79.5 9.5
Average 53733 66967 13233 0.80 31.2 262.5 78.7 9.3

0.625 Mean 53844 66656 12811 0.81 31.8 262.5 77.5 9.3
Standard Dev. 1051 510 990 0.01 1.0 0.0 1.2 0.5
Note 1. Machine capacity was exceeded on 0.625” CVN tests and the samples did not break, max machine value is listed.

9
Pipelines 2010: Climbing New Peaks to Infrastructure Reliability—Renew, Rehab, and Reinvest 390
© 2010 ASCE

Table 3 Test Results for 0.207” thickness 30” Diameter Gasket Bell Pipe
CVN
Yield Tensile CVN CVN Corrected
Thickness Diameter Sample psi psi AVG Size Sub- For Sub- Rockwell Grain
in. in. Location (Y) (T) T-Y Y/T Elongation Ft-lbs mm size size B Avg Size
0.207 30 Pipe Body 56400 75900 19500 0.74 29.0 37.8 5 Y 75.7 80.5 10.5
0.207 30 Pipe Body 54300 76200 21900 0.71 28.0 37.0 5 Y 74.0 82.0 10.5
Average 55350 76050 20700 0.73 28.5 37.4 81.3 10.5

Pipe Body
0.207 30 Hoop 53200 76200 23000 0.70 29.5 40.0 5 Y 80.0 83.5 10.5
Pipe Body
0.207 30 Hoop 55000 75800 20800 0.73 27.0 40.0 5 Y 80.0 81.5 10.5
Average 54100 76000 21900 0.71 28.3 40.0 82.5 10.5

Gasket
0.207 30 Bell 56200 76100 19900 0.74 28.0 39.2 5 Y 78.3 83.5 10.5
Gasket
0.207 30 Bell 50300 76300 26000 0.66 28.5 41.2 5 Y 82.3 84.0 10.5
Average 53250 76200 22950 0.70 28.3 40.2 83.8 10.5

Rolled
0.207 30 Spigot 54800 74700 19900 0.73 26.5 40.2 5 Y 80.3 81.5 10.5
Rolled
0.207 30 Spigot 50900 75900 25000 0.67 28.0 39.5 5 Y 79.0 80.0 10.5
Average 52850 75300 22450 0.70 27.3 39.8 80.8 10.5

0.207
Mean 53888 75888 22000 0.71 28.1 39.0 82.1 10.5
Standard
Dev 2126 478 2305 0.03 0.9 2.6 1.4 0

10
Pipelines 2010: Climbing New Peaks to Infrastructure Reliability—Renew, Rehab, and Reinvest 391
© 2010 ASCE

Table 4 Test Results for 0.149” thickness 30” Diameter Gasket Bell Pipe
CVN
Yield Tensile CVN CVN Corrected
Thickness Diameter Sample psi psi AVG Size Sub- For Sub- Rockwell Grain
in. in. Location (Y) (T) T-Y Y/T Elongation Ft-lbs mm size size B Avg Size
0.149 30 Pipe Body 50400 72100 21700 0.70 29.0 12.0 2.5 Y 48.0 81.0 10.5
0.149 30 Pipe Body 53800 72900 19100 0.74 28.0 13.7 2.5 Y 54.7 80.0 10.5
Average 52100 72500 20400 0.72 28.5 12.8 80.5 10.5

Pipe Body
0.149 30 Hoop 52900 71800 18900 0.74 29.0 11.2 2.5 Y 44.7 80.0 10.5
Pipe Body
0.149 30 Hoop 52800 73300 20500 0.72 30.0 14.0 2.5 Y 56.0 82.0 10.5
Average 52850 72550 19700 0.73 29.5 12.6 81.0 10.5

0.149 30 Gasket Bell 55400 73000 17600 0.76 30.5 13.3 2.5 Y 53.3 81.5 10.5
0.149 30 Gasket Bell 58600 76400 17800 0.77 24.0 13.7 2.5 Y 54.7 86.5 10.5
Average 57000 74700 17700 0.76 27.3 13.5 84.0 10.5

0.149 30 Rolled Spigot 54500 72900 18400 0.75 28.5 13.5 2.5 Y 54.0 80.0 10.5
0.149 30 Rolled Spigot 53500 72800 19300 0.73 29.0 13.7 2.5 Y 54.7 80.5 10.5
Average 54000 72850 18850 0.74 28.8 13.6 80.3 10.5

0.149 Mean 53988 73150 19163 0.74 28.5 13.1 81.4 10.5
Standard
Dev. 2216 1312 1284 0.02 1.9 0.9 2.0 0

11
Pipelines 2010: Climbing New Peaks to Infrastructure Reliability—Renew, Rehab, and Reinvest 392
© 2010 ASCE

References

American Water Works Association (AWWA) (2007) C200-05 Steel Water

Pipe—6 In. (150 mm) and Larger, Denver, CO

ASTM (2008) E18 - 08b Standard Test Methods for Rockwell Hardness of
Metallic Materials, West Conshohocken, PA

ASTM (2004) E112 – 96 (2004) e2 Standard Test Methods for Determining

Average Grain Size, West Conshohocken, PA

ASTM (2009) A370 - 09a Standard Test Methods and Definitions for Mechanical
Testing of Steel Products, West Conshohocken, PA

ASTM (2009) A1011/ A1011M - 09b Standard Specification for Steel, Sheet and
Strip, Hot-Rolled, Carbon, Structural, High-Strength Low-Alloy, High-Strength
Low-Alloy with Improved Formability, and Ultra-High Strength, West
Conshohocken, PA

ASTM (2009) A1018 / A1018M - 09 Standard Specification for Steel, Sheet and
Strip, Heavy-Thickness Coils, Hot-Rolled, Carbon, Commercial, Drawing,
Structural, High-Strength Low-Alloy, High-Strength Low-Alloy with Improved
Formability, and Ultra-High Strength, West Conshohocken, PA

Maxey, W.A., Eiber, R.J. (1988) A Study of the Yield/Tensile Ratio and its Effect
on Line Pipe Behavior, Battelle Memorial Institute

Shoemaker, A.K. (1984) The Effect of Plate Stress-Strain Behavior and

Pipemaking Variables on the Yield Strength of Large-Diameter DSAW Line Pipe,
Journal of Engineering Materials and Technology, April 1984, ASME

12