Proceedings

Proceedings of IPC’02
of IPC 2002:
th
4th 4International
International Pipeline
Pipeline Conference
Conference 2002
September 29-October 3, 2002, Calgary, Alberta,
September 29-October 3, 2002, Calgary, Alberta, Canada Canada

IPC2002-27131
IPC02-27131

QUALIFICATION OF PROCEDURES FOR

WELDING ONTO IN-SERVICE PIPELINES
William A. Bruce
Materials Section, Edison Welding Institute

ABSTRACT INTRODUCTION
When welding onto an in-service pipeline, to facilitate a The risk of burnthrough is negligible if the wall thickness
repair or to install a branch connection using the "hot tapping" is 0.250 in. (6.4 mm) or greater, provided that low hydrogen
technique, two risks need to be considered. The first is the risk (EXX18 type) electrodes and normal welding practices are
of burnthrough, where the welding arc causes the pipe wall to used [1]. Welding onto thinner-wall in-service pipelines is
be penetrated allowing the contents to escape. The second is possible and considered routine by many companies; however,
the risk of hydrogen cracking that arises from the fast cooling special precautions, such as the use of a procedure that limits
rates that tend to be produced by the ability of the flowing heat input, are often specified. Hydrogen cracking requires that
contents to remove heat from the pipe wall. To prevent three conditions be fulfilled simultaneously: hydrogen in the
hydrogen cracking, at least one of the three conditions weld, a crack-susceptible weld microstructure, and stress acting
necessary for its occurrence must be eliminated. Beyond the on the weld. To prevent hydrogen cracking, at least one of the
use of low-hydrogen electrodes to minimize hydrogen levels, it three conditions necessary for its occurrence must be
is prudent to develop and use procedures that minimize the eliminated. As the result of several significant incidents that
formation of crack susceptible microstructures. This paper have occurred in the past, it has become common for in-service
reviews existing methods for selecting parameters and welds to be made using low-hydrogen electrodes. Since low
qualifying procedures for welding onto in-service pipelines. hydrogen levels cannot always be guaranteed, many companies
have developed procedures that minimize the formation of
HAZ hardness is an indicator of the susceptibility of a crack-susceptible microstructures as an additional protection
microstructure to cracking. A widely-used value below which against hydrogen cracking. Procedure options for minimizing
it is generally agreed that hydrogen cracking is not expected is the formation of crack susceptible microstructures for welds
350 HV. Unfortunately, there is no one hardness level above made onto in-service pipeline include the use of a sufficiently-
which the risk of hydrogen cracking becomes unacceptable. high heat input level, the use of preheating, the use of a temper
This paper also describes the development of a hardness bead deposition sequence, or some combination of these.
evaluation criterion that can be used to quantify the trade-offs
that can be made between HAZ hardness, hydrogen level, and METHODS FOR PREDICTING REQUIRED WELDING
chemical composition. PARAMETERS AND QUALIFYING PROCEDURES
There are three commonly-used methods for predicting
Finally, the results of a recently-completed group- welding parameters required to avoid hydrogen cracking for
sponsored project, where procedures for welding onto in- welds made onto in-service pipelines; thermal analysis
service pipelines were developed over a wide range of modeling, heat sink capacity measurement, and iterative
conditions, are also reviewed. The results can be used to select procedure qualification trials. Either of the first two methods
an appropriate procedure that is resistant to hydrogen cracking can be used to establish a starting point for the third method.
for a particular application. The use of these results allows in-
service welding to be carried out in a safe, cost-effective Predicting Required Heat Input Levels
manner, allowing both economic and environmental benefits to The most common procedures for minimizing the formation
be realized by avoiding pipeline shutdown and interruption of of crack-susceptible microstructures use a sufficiently high heat
service. input level to overcome the effect of the flowing contents.
Thermal analysis modeling can be used for predicting required
heat input levels. The most commonly used model was

1 Copyright © 2002 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/pdfaccess.ashx?url=/data/conferences/ipc2002/69711/ on 02/23/2017 Terms of Use: http://www.asme.org/ab

developed by Battelle [2,3]. The Battelle model uses two- The purpose of qualifying a welding procedure is to
dimensional numerical solutions of heat transfer equations to demonstrate that the procedure is capable of producing sound
predict inside surface temperatures and cooling rates for single- welds under production conditions. The purpose of a welder
pass fillet welds at the end of a sleeve or a branch-to-carrier qualification is to show that a particular welder is capable of
pipe groove weld. executing a qualified procedure under production conditions. It
is clearly unrepresentative to qualify procedures and welders
A second method for predicting required heat input levels for welding onto in-service pipelines using a length of pipe that
relies on heat sink capacity measurement. This method was contains still air. Without simulating the ability of the in-
developed at EWI [3] and involves measuring the ability of the service pipeline to remove heat from the pipe wall,
flowing contents to remove heat from the pipe wall using a unrealistically slow weld cooling rates result, in addition to
simple field test (Figure 1). This test involves quickly heating different solidification characteristics of the weld pool.
a 2-in. (50-mm) diameter area on the pipeline with an oxy-fuel
torch to between 300 and 325ºC. The time required for the area Several approaches exist for qualifying procedures and
to cool from 250 to 100ºC is then measured using a digital welders where the thermal conditions experienced when
contact thermometer and a stopwatch. The measured heat-sink welding onto an actual in-service pipeline are simulated. The
capacity value is used to predict the weld cooling rates using results of the PRCI-sponsored work at EWI [6] show that the
empirical relationships that were developed from data use of water as a flow medium is more severe with respect to
generated in the field and in the laboratory for a wide range of the resulting weld cooling rates than with most hydrocarbon
conditions. liquids and high-pressure gasses. The use of water also reduces
the complications associated with handling hazardous liquids
Both of these methods allow welding parameters (i.e., heat (Figure 2). The results also show that the use of the motor oil
input levels) to be selected based on anticipated weld cooling as a flow medium is appropriate for less severe conditions (e.g.,
rates. Limits on the weld cooling rates are established based on low-pressure, thinner-wall natural gas pipelines operating at
the maximum tolerable HAZ hardness predicted using a low flow rates). The experimental data can be used to establish
previously-established empirical correlation [4] and the the limiting flow conditions for procedures qualified under
anticipated carbon equivalent of the pipe material. This each of the procedure qualification conditions.
previously-established empirical correlation aims to limit HAZ
hardness to 350 HV. Hardness Evaluation Criterion
Both the Battelle model and the EWI method use a critical
EWI recently completed the development of an improved hardness level (i.e., the hardness level below which these welds
thermal analysis model for PRCI [5]. The model uses a were considered acceptable) of 350 HV. In previous work [7],
proprietary finite-element solver that was developed at EWI. the critical hardness level has been shown to depend on the
Mesh generation capabilities include sleeve, branch, and bead- hydrogen level of the welding process being used and on the
on-pipe geometries (the latter for buttering layers and weld chemical composition (carbon equivalent or CE level) of the
deposition repairs). The user interface allows multiple cases to materials being welded. Closer control of hydrogen level
be run and enhanced heat input selection curves to be allows higher hardness to be tolerated. Since the hardness of
generated. The model allows welding parameters (i.e., heat martensite depends on the carbon level, lower CE materials
input levels) to be selected that limit HAZ hardness to a user- tend to crack at lower hardness levels. Therefore higher
selected value. hardness can be tolerated when welding higher CE materials.

Procedure Qualification Methods Critical hardness level is shown as a function of CEIIW

Both thermal analysis modeling and heat sink capacity level from several previous programs in Figure 3. The data
measurement allow required welding parameters to be shown in this figure are for fillet welds made at a hydrogen
predicted based on anticipated weld cooling rates. These level of approximately 12 ml/100 gm. In previous work at
predictive methods do not guarantee that these parameters are EWI [8], this figure was used as a starting point for the
practical under field conditions, however, and neither is development of an evaluation criterion for welds made onto in-
capable of predicting the effect of tempering from subsequent service pipelines. This hardness evaluation criterion was
passes for a temper bead procedure. To demonstrate that the developed to quantify the trade-offs that can be made between
parameters are practical, a welding procedure based on hardness, hydrogen level, and chemical composition.
predicted parameters should be qualified under simulated
conditions. Simple methods have been developed for An evaluation criterion that is loosely based on Figure 3 is
establishing procedure qualification conditions that simulate shown in Figure 4. For each CE level, the critical hardness
the thermal conditions experienced when welding onto an level for the 8 ml/100 gm hydrogen level is roughly that
actual in-service pipeline (i.e., that result in realistic weld extrapolated from a lower bound of the data in Figure 3. A 25
cooling rates and solidification characteristics). HV increase in the critical hardness was allowed for the 4

2 Copyright © 2002 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/pdfaccess.ashx?url=/data/conferences/ipc2002/69711/ on 02/23/2017 Terms of Use: http://www.asme.org/ab

ml/100 gm hydrogen level. A 50 HV decrease was imposed for testing. Following the laboratory-scale trials, full-scale
the welds made using cellulosic-coated electrodes. procedure qualification trials were carried out to demonstrate
that the procedures are capable of being executed under
To evaluate this criterion, highly-restrained multi-pass production conditions.
fillet welds were made under simulated in-service conditions
(Figure 5) on materials with a variety of chemical compositions SELECTING AN APPROPRIATE PROCEDURE
(Table 1) using a variety of different hydrogen levels, after None of the commonly-used procedure options -- the use
which they were metallography sectioned to measure hardness of a sufficiently-high heat input level, the use of preheating, or
at the weld toe and to look for cracks. Comparison of the the use of a temper bead deposition sequence -- are entirely
hardness testing results with the evaluation criterion (Figure 6) straightforward. Maintaining heat input above specified
indicates that the criterion has accurately predicted the minimum required levels can be difficult to monitor, and field
occurrence of cracking (i.e., welds that are cracked tend not to control can be difficult. Excessive heat input levels may result
meet the criterion). The criterion is useful in determining how in an increased risk of burnthrough, particularly for thin wall
close the use of a certain procedure in a specific application is pipe. Under certain severe conditions, even high heat input
to the cracking threshold. welds can cool faster than the threshold value required to avoid
the formation of a hard, crack-susceptible microstructure.
While this criterion has been effective for the conditions Under these severe conditions it is often a better approach to
shown here (%C>0.10, 0.375 in. thick materials, etc.), use a temper-bead procedure.
additional work is needed in this area.
Temper-bead procedures rely on heat from subsequent
DEVELOPMENT AND QUALIFICATION OF passes to refine and temper the heat-affected zone (HAZ) of
PROCEDURES previous passes (Figure 7). These procedures require
Using the methods for predicting required welding considerable skill on the part of the welder, and are not
parameters and the procedure qualification methods described necessary for less unfavorable conditions, such as for low
above, EWI conducted a group-sponsored project (GSP) to carbon equivalent materials and less severe thermal conditions.
develop procedures for welding onto in-service pipelines that Also, unscheduled departures from the temper-bead procedure
are resistant to hydrogen cracking over a wide range of can result in higher finished-weld hardness levels than if a
conditions. These procedures were qualified to the standard procedure had been used. For less unfavorable
requirements of a variety of industry codes, on behalf of an conditions, it is often a better approach to use a standard multi-
international group of sponsors. pass procedure. Welds with unfavorable profile have been
known to result from the use of temper-bead procedures, since
In the GSP, the procedures that were developed rely on weld bead placement is prescribed by tempering requirements
heat input control (15, 25, and 40 kJ/in. [0.6, 1.0, and 1.6 and cannot be changed to suit the requirements for proper
kJ/mm]), preheat temperature control (200°F [93°C]), a temper filling of each individual weld.
bead deposition sequence, or some combination of these. A
procedure that uses austenitic (stainless steel) weld metal, For procedures based on preheat control, maintaining
which acts as a hydrogen sponge, was also developed. The specified preheat levels is often difficult, particularly for thin
heat input control procedures were qualified using both wall pipe. If non-continuous preheating methods are used, the
shielded metal arc welding (SMAW) and gas metal arc welding re-heating interval may be very short. Monitoring of preheat
(GMAW). The development matrix includes three hydrogen levels in the field can also be difficult.
levels (4 and 8 ml/100 gm for low hydrogen electrodes, and a
third hydrogen level for cellulosic-coated electrodes), two wall It can be shown that, for some in-service welding
thickness groups (0.375 and 0.750 in. [9.5 and 19.1 mm]), two applications, even the most simple procedures (e.g., standard
thermal severity levels (simulated using either flowing water or fillet welds) result in acceptable hardness levels. This approach
flowing motor oil), and four pipe material chemical is most attractive in that simple procedures are the least
composition groups (0.35, 0.42, and 0.50 CEIIW for expensive to execute and involve fewer opportunities for
conventional steels, and a modern composition with a carbon errors. The most cost-effective approach to welding onto in-
content of less than 0.10%). Details of the procedures that service pipelines is to have a standard set of procedures
were developed are shown in Table 2. qualified ranging in complexity, and then to specify the least
complex procedure required for each application. The
Laboratory-scale procedure development and qualification objective of the recently-completed group-sponsored project at
trials were carried out to demonstrate the ability of the EWI [8] addressed this need. Procedures were developed using
procedures to produce sound, crack-free welds over a range of the options described above and then a guideline was
simulated conditions. Evaluation of these welds included developed that allows the least complicated procedure to be
mechanical testing, metallographic examination, and hardness selected for a specific application.

3 Copyright © 2002 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/pdfaccess.ashx?url=/data/conferences/ipc2002/69711/ on 02/23/2017 Terms of Use: http://www.asme.org/ab

Required Parameters To use the methane flow rate method, the normalized
The purpose of this section is to guide the user through the volumetric flow rate is calculated by dividing the volumetric
process of selecting an appropriate procedure for welding onto flow rate (in mmscfd) by the cross-sectional area (in square
in-service pipelines, based on the results of the recently- inches) of the inside diameter of the pipe. Figure 9 can be used
completed GSP [8]. The procedure selection guidelines are to determine the thermal severity level by locating the point of
shown in the form of a flow chart in Figure 8. The intersection of the volumetric flow rate and the pipe wall
corresponding procedure variations are shown in Table 2. The thickness. In general, the low thermal-severity category
use of this figure requires knowledge of the hydrogen level of (Category II) will apply only to thinner-wall pipelines
the welding process, the pipe wall thickness, the thermal transporting methane at low flow rates. As an alternative to the
severity level (for materials 0.5-in. [12.7-mm] thick and less), use of Figure 9, an estimate of the maximum allowable flow
and the pipe material chemical composition. Where any one of rate for Category II thermal severity level can be made as
these items of information is unavailable, a conservative follows:
estimate must be made. The following sections provide
guidance pertaining to the determination of these parameters. FMAX = [3.4 – 10t] A

Hydrogen Level. Each of the three charts in of Figure 8 Where:

is intended for one of three hydrogen levels: <4 ml/100 gm, <8 FMAX = maximum allowable flow rate (mmscfd)
ml/100 gm, and that produced by cellulosic-coated electrodes. t = wall thickness (in.)
Chart 1 (<4 ml/100 gm) is intended for applications where low- A = area of pipe inside diameter (in.2)
hydrogen electrodes or the GMAW process will be used and
the conditions are favorable in terms of moisture and To use the heat sink capacity method, the heat sink
contamination on the consumables and on the pipe and fitting capacity measurement procedure described above must be
surfaces. To ensure that low-hydrogen electrodes are clean and carried out. Knowing the results of this measurement and the
dry, they should be used directly out of a freshly-opened can, or pipe wall thickness, Figure 10 can be used to determine the
directly from an electrode holding oven. Alternatively, thermal severity level.
specially-packaged low-hydrogen electrodes (e.g., packaged in
Sahara ReadyPacks, etc.) can be used and are ideally suited for The user needs to know the thermal severity level to select
this application. the procedure variation, but also needs to know the procedure
variation (i.e., the heat input) to know which curve to use in
Chart 2 (<8 ml/100 gm) is intended for applications where Figures 9 and 10 to determine the thermal severity level. To
low-hydrogen electrodes cannot be cared for as described alleviate this circular reference, an assumption must be made
above, such as applications where electrode holding ovens are initially and rechecked following the procedure selection
not available, or where conditions are unfavorable in terms of process. It is generally safe to use the curves for a heat input of
moisture and contamination. The latter pertains to both SMAW 40 kJ/in. as a starting point.
and GMAW. An example of an unfavorable condition would
include atmospheric conditions that result in sweating of the Pipe Material Chemical Composition. In terms of
pipeline which cannot be controlled by preheating. chemical composition, procedure selection using the guideline
should be made based on the least favorable of the materials
Chart 3 is intended for applications where cellulosic- being welded (e.g., the pipe material or the fitting material,
coated electrodes, which produce a hydrogen level in the 40-60 whichever has the highest carbon equivalent level). Where
ml/100 gm range, are to be used. records that contain chemical composition information are
unavailable, a small sample of filings can be removed for
Thickness. In terms of wall thickness, selection of a laboratory analysis. The two required parameters are %C and
procedure using the guideline should be made based on the the carbon equivalent, calculated using the IIW formula, as
thickness of the pipe material, since this is most influential with follows:
regard to the resulting weld cooling rate.
CEIIW = %C + %Mn/6 + (%Cu + %Ni)/15 + (%Cr + %Mo
Thermal Severity Level. For materials <0.5-in. thick, + %V)/5
the use of the guideline requires that the thermal severity level
be determined: Category I for high thermal severity (fast weld Unfortunately, records that contain chemical composition
cooling rates), and Category II for low thermal severity (slow information for older, existing pipelines are usually difficult or
weld cooling rates). If the pipe content is methane, this can be impossible to locate. The use of chemical composition
accomplished using either the flow rate method or the heat sink estimates that are based on the maximum allowable limits of
capacity method. For other pipe contents, the heat sink the specification to which the materials were produced can be
capacity method must be used. restrictively over-conservative, and direct analysis using

4 Copyright © 2002 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/pdfaccess.ashx?url=/data/conferences/ipc2002/69711/ on 02/23/2017 Terms of Use: http://www.asme.org/ab

portable equipment is cumbersome and expensive and requires unstable during welding (e.g., ethylene). Once an appropriate
a significant calibration and maintenance effort. procedure has been selected for the application of interest, if
the wall thickness is less than 0.250 in., the risk of burnthrough
Chemical composition determination by removal of should be independently evaluated. Similarly, if there is a risk
samples for laboratory analysis is relatively simple and can be of the pipeline contents becoming unstable, this too must be
quite accurate, provided that care is taken in removing the independently evaluated. The most useful tool in evaluating
samples and that they are properly analyzed. The most this risk is a thermal analysis computer model that was
effective method of sample removal involves the use of a high- developed by Battelle [2,3].
speed rotary file (burr grinder). The wall thickness remaining
after sample removal should be greater than the specified Availability of Qualified Procedures
nominal wall thickness minus the pipe mill tolerance, which The original sponsor group for this project, which consists
can be up to 12.5%. The rotary file should be inspected of nine companies, elected to release the results of the project,
following use to ensure that teeth have not broken off, as the of which the guidelines described above are a part, although the
inclusion of teeth in the filings can raise the apparent content of qualified procedures remain proprietary. The sponsor group
certain elements, particularly carbon. The filings can be also elected to make the full results available to additional
analyzed using traditional wet chemistry (i.e., titration companies through a reduced rate offer [9]. The revenue
methods) or using the inductively coupled plasma (ICP) generated by this offer is being used to conduct further work in
method. The combustion-in-oxygen method using a LECO this area. To date, a total of 18 companies have joined the
analyzer can be used to accurately determine carbon and sulfur project.
levels of samples composed of filings.
The final report for this project also contains guidance
Using the Procedure Selection Guidelines pertaining to a number of aspects of applying qualified
Once the required parameters have been determined for a procedures in the field. These include handling of low-
given application, the procedure selection process begins by hydrogen electrodes, control of heat input levels, avoidance of
selecting either Chart 1, 2, or 3 based on the hydrogen level of welder-induced discontinuities, inspection of completed
the welding process that will be used. Each chart is then repairs, and repair and removal of defects.
divided into sections for thin (<0.5-in.) and thick (>0.5-in.)
pipe wall thickness. For pipe <0.5-in. thick, each chart is then SUMMARY
divided into high (Category I) and low (Category II) thermal When welding onto an in-service pipeline, there is a risk of
severity levels. Each section is then divided into four chemical hydrogen cracking that arises from the fast cooling rates that
composition categories: one for pipe material with <0.10%C tend to be produced by the ability of the flowing contents to
and three for pipe material with >0.10%C. For material with remove heat from the pipe wall. There are a variety of methods
>0.10%C, there are three carbon equivalent (CEIIW) categories: that exist for predicting welding parameters required to avoid
<0.35, <0.42, and <0.50. hydrogen cracking and for qualifying procedures for welding
onto in-service pipelines. These methods were used during a
Once the user finds the appropriate column for the recently-completed group-sponsored project, where procedures
application of interest, the required procedure option is then that are resistant to hydrogen cracking were developed for a
selected from the list. All of the procedure options contained in wide range of in-service welding applications. The results of
the list are suitable, although the more-simple procedures are this project allows the least complicated procedure to be
contained at the top of each list. If there is no suitable selected for a specific application. The use of these results
procedure listed in the appropriate column, then the starting allows in-service welding to be carried out in a safe, cost-
parameters must be changed (e.g., the hydrogen level or the effective manner, allowing both economic and environmental
thermal severity level must be reduced) and the process benefits to be realized by avoiding pipeline shutdown and
repeated. interruption of service.

As noted in Charts 2 and 3, weld metal hydrogen cracks REFERENCES

tended to be produced if hydrogen levels were not carefully 1. Kiefner, J. F. and Fischer, R. D., "Repair and Hot Tap
controlled. This was particularly true for welds made at the Welding on Pressurized Pipelines," Symposium during
high thermal severity level. The significance of these cracks, 11th Annual Energy Sources Technology Conference and
which tended to be small and discontinuous, was not Exhibition, New Orleans, LA, January 10-13, 1988, (New
determined. York, NY: American Society of Mechanical Engineers,
PD-Vol. 14., 1987) pp. 1-10.
It should be noted that the GSP did not specifically address 2. Bubenik, T A, Fischer, R D, Whitacre, G R, Jones, D J,
general safety-during-welding issues (e.g., burnthrough) or Kiefner, J F, Cola, M J, and Bruce, W A, 1991.
concerns related to pipeline products which may become Investigation and prediction of cooling rates during

5 Copyright © 2002 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/pdfaccess.ashx?url=/data/conferences/ipc2002/69711/ on 02/23/2017 Terms of Use: http://www.asme.org/ab

pipeline maintenance welding, Final Report to American 6. Bruce, W A and Threadgill, P L 1994. Evaluation of the
Petroleum Institute. effect of procedure qualification variables for welding onto
3. Cola, M J, Kiefner, J F, Fischer, R D, Jones, D J, and in-service pipelines, Final Report to A.G.A. Pipeline
Bruce, W A 1992. Development of simplified weld cooling Research Committee for PR-185-9329, EWI Project No.
rate models for in-service gas pipelines, Project Report No. J7141, Edison Welding Institute, Columbus, OH.
J7134 to A.G.A. Pipeline Research Committee, Edison 7. Hart, P H M and Matharu, I S 1985. Heat-affected zone
Welding Institute, Kiefner and Associates and Battelle (HAZ) hydrogen cracking behavior of low carbon
Columbus Division, Columbus, OH. equivalent C-Mn structural steels, TWI Research Report
4. Graville, B A and Read, J A, 1974. Optimization of fillet No. 290/1985.
weld sizes, Welding Research Supplement, Welding 8. Bruce, W A 1996. “Qualification and selection of
Journal. procedures for welding onto in-service pipelines and
5. Bruce, W A, Li, V, Citterberg, R, Wang, Y-Y, and Chen, Y. piping systems, Final Report to a Group of Sponsors,
2001. Improved cooling rate model for welding on in- Edison Welding Institute, Columbus, OH.
service pipelines, PRCI Contract No. PR-185-9633, EWI 9. Procedures for welding onto in-service pipelines, Group-
Project No. 42508CAP, Edison Welding Institute, Sponsored Project Brochure, Edison Welding Institute,
Columbus, OH Columbus, OH, 1996.

6 Copyright © 2002 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/pdfaccess.ashx?url=/data/conferences/ipc2002/69711/ on 02/23/2017 Terms of Use: http://www.asme.org/ab

Table 1. Chemical Composition of Materials used to Evaluate Hardness Acceptance Criterion

Chemical Composition, wt.%

Element N2 N3 N4
C 0.14 0.24 0.38
Mn 1.14 1.10 0.79
P 0.010 0.016 0.016
S 0.001 0.014 0.002
Si 0.33 0.19 0.22
Ni 0.024 0.020 0.018
Cr 0.037 0.040 0.033
Mo 0.013 0.001 0.013
Cu 0.011 0.020 0.023
V 0.001 0.004 0.000
Al 0.024 0.047 0.043
Ti 0.000 0.000 0.004
Nb 0.000 0.002 0.000
Zr 0.000 0.000 0.000
W 0.00 0.00 0.00
Pb 0.00 0.00 0.00
B 0.0000 0.0000 0.0002
CEIIW 0.34 0.44 0.52
Pcm 0.21 0.31 0.43

Table 2. Details of Procedures that were Developed

Procedure Hydrogen Cracking Required

Variation Control Method Parameters
A Heat input control 15 kJ/in. (0.6 kJ/mm)
minimum required heat input
B Heat input control 25 kJ/in. (1.0 kJ/mm)
minimum required heat input
C Heat input control 40 kJ/in (1.6 kJ/mm)
minimum required heat input
D1 Temper bead sequence 15 kJ/in. (0.6 kJ/mm) first layer,
25 kJ/in. (1.0 kJ/mm) second layer
D2 Austenitic weld metal As required

E Heat input control w/ preheat 25 kJ/in. (1.0 kJ/mm)

w/200ºF (93ºC) preheat
F Heat input control w/ preheat 40 kJ/in (1.6 kJ/mm)
w/200ºF (93ºC) preheat

7 Copyright © 2002 by ASME 7 7 Copyright © 2002 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/pdfaccess.ashx?url=/data/conferences/ipc2002/69711/ on 02/23/2017 Terms of Use: http://www.asme.org/ab

Figure 1. Heat Sink Capacity Measurement

Figure 2. Procedure Qualification Set-Up Using Flowing Water

Copyright © 2002 by ASME 8 8 Copyright © 2002 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/pdfaccess.ashx?url=/data/conferences/ipc2002/69711/ on 02/23/2017 Terms of Use: http://www.asme.org/ab

Fillet Welds Made
at Hydrogen Level
of ~12 ml/100 gm

Figure 3. Critical Hardness Level vs. CEIIW Level from Several Previous Programs

Materials with %C > 0.10

450
Low hydrogen - < 4 ml/100
Low hydrogen - < 8 ml/100
Cellulosic-coated
400
Critical HAZ hardness, HV

350

300

250
0.2 0.3 0.4 0.5 0.6
Carbon equivalent, CEIIW

Figure 4. Critical Hardness Level for In-Service Welds vs. CEIIW and Weld Hydrogen Level

Copyright © 2002 by ASME 9 9 Copyright © 2002 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/pdfaccess.ashx?url=/data/conferences/ipc2002/69711/ on 02/23/2017 Terms of Use: http://www.asme.org/ab

Figure 5. Set-Up for Simulated Sleeve Fillet Welding Trials

0.375 in. (9.5 mm) Thick Materials with %C > 0.10

550
Measured Average HAZ HArdness, HV-10kg

500 Non-Cracked
Cracked
450 1:1

400

350

300

250

200

150
150 200 250 300 350 400 450 500 550
Maximum-Allowable HAZ Hardness, HV

Figure 6. Comparison of the Hardness Testing Results with Hardness Evaluation Criterion

Copyright © 2002 by ASME 1010 Copyright © 2002 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/pdfaccess.ashx?url=/data/conferences/ipc2002/69711/ on 02/23/2017 Terms of Use: http://www.asme.org/ab

Figure 7. Illustration of a Temper Bead Deposition Sequence

Low Hydrogen
(EXX18-type), W hat
Cellulosic-coated
GMAW , or electrodes/process
(EXX10-type)
Austenitic will be used?
(E309-type)

Are conditions favorable for Are conditions favorable for

No
hydrogen levels<=4mL/100gm? hydrogen levels <=8m l/100gm?

No

Yes Yes

Use Chart 1 Use Chart 2 Use Chart 3

Figure 8a. Procedure Selection Guideline for Welding onto In-Service Pipelines (cont.)

Copyright © 2002 by ASME 1111 Copyright © 2002 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/pdfaccess.ashx?url=/data/conferences/ipc2002/69711/ on 02/23/2017 Terms of Use: http://www.asme.org/ab

Chart 1

W hat is the wall

<=0.5" >0.5"
thickness?

W hat is the thermal

severity?
Category I (fast cooling) Category II (slow cooling)

W hat is the %C? W hat is the %C? W hat is the %C?

>0.10% >0.10% >0.10%

<=0.10% <=0.10% <=0.10%

W hat is the CE IIW ? W hat is the CE IIW ? What is the CE IIW ?

<=0.35% <=0.42% <=0.50% <=0.35% <=0.42% <=0.50% <=0.35% <=0.42% <=0.50%

Acceptable Acceptable Acceptable Acceptable Acceptable Acceptable Acceptable Acceptable Acceptable Acceptable Acceptable Acceptable
procedure procedure procedure procedure procedure procedure procedure procedure procedure procedure procedure procedure
variations: variations: variations: variations: variations: variations: variations: variations: variations: variations: variations: variations:
A A A A A
B B B B B B B B
C C C C C C C E/C E/C E/C E/C
D1/D2 D1/D2 D1/D2 D1/D2 D1/D2 D1/D2 D1/D2 D1/D2 F/D1/D2 F/D1/D2 F/D1/D2 F/D1/D2

See Note 2 See Note 2 See Note 2 See Note 2 See Note 2 See Note 2 See Note 2 See Note 2

Note 1 - Reduce hydrogen level or thermal severity level

Note 2 - If wall thickness is <0.250" check burnthrough risk using Battelle Model.

Figure 8b. Procedure Selection Guideline - Chart 1 (<4 ml/100 gm hydrogen level)

12 Copyright © 2002 by ASME

12 12
Copyright © 2002 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/pdfaccess.ashx?url=/data/conferences/ipc2002/69711/ on 02/23/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use

Chart 2

W hat is the wall

<=0.5" >0.5"
thickness?

W hat is the thermal

severity?
Category I (fast cooling) Category II (slow cooling)

W hat is the %C? W hat is the %C? W hat is the %C?

>0.10% >0.10% >0.10%

<=0.10% <=0.10% <=0.10%

W hat is the CE IIW ? W hat is the CE IIW ? What is the CE IIW ?

<=0.35% <=0.42% <=0.50% <=0.35% <=0.42% <=0.50% <=0.35% <=0.42% <=0.50%

Acceptable Acceptable Acceptable Acceptable Acceptable Acceptable Acceptable Acceptable Acceptable Acceptable Acceptable Acceptable
procedure procedure procedure procedure procedure procedure procedure procedure procedure procedure procedure procedure
variations: variations: variations: variations: variations: variations: variations: variations: variations: variations: variations: variations:
A A
B B B
C* C C C C E*/C* E*/C*
D1* D1* D1* D1 D1 D1 D1 F*/D1* F*/D1*
See Note 1 See Note 1 See Note 1
See Note 2 See Note 2 See Note 2 See Note 2 See Note 2 See Note 2 See Note 2

Note 1 - Reduce hydrogen level or thermal severity level

Note 2 - If wall thickness is <0.250" check burnthrough risk using Battelle Model.
* These procedure variations have a propensity to produce weld metal hydrogen cracking.

Figure 8c. Procedure Selection Guideline - Chart 2 (<8 ml/100 gm hydrogen level)

13 Copyright © 2002 by ASME 13 Copyright © 2002 by ASME

13

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/pdfaccess.ashx?url=/data/conferences/ipc2002/69711/ on 02/23/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use

Chart 3

W hat is the wall

<=0.5" >0.5"
thickness?

W hat is the thermal

severity?
Category I (fast cooling) Category II (slow cooling)

W hat is the %C? W hat is the %C? W hat is the %C?

>0.10% >0.10% >0.10%

<=0.10% <=0.10% <=0.10%

W hat is the CE IIW ? W hat is the CE IIW ? What is the CE IIW ?

<=0.35% <=0.42% <=0.50% <=0.35% <=0.42% <=0.50% <=0.35% <=0.42% <=0.50%

Acceptable Acceptable Acceptable Acceptable Acceptable Acceptable Acceptable Acceptable Acceptable Acceptable Acceptable Acceptable
procedure procedure procedure procedure procedure procedure procedure procedure procedure procedure procedure procedure
variations: variations: variations: variations: variations: variations: variations: variations: variations: variations: variations: variations:

B*
C* C* C* C*
D1* D1* D1* F* F*
See Note 1 See Note 1 See Note 1 See Note 1 See Note 1
See Note 2 See Note 2 See Note 2 See Note 2 See Note 2

Note 1 - Reduce hydrogen level or thermal severity level

Note 2 - If wall thickness is <0.250" check burnthrough risk using Battelle Model.
* These procedure variations have a propensity to produce weld metal hydrogen cracking.

Figure 8d. Procedure Selection Guideline - Chart 3 (cellulosic-coated electrodes)

14 Copyright © 2002 by ASME 14 Copyright © 2002 by ASME

14

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/pdfaccess.ashx?url=/data/conferences/ipc2002/69711/ on 02/23/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use

1.6
Normalized volumetric flow rate, mmscfd/sq. in.
1.4

1.2
Category I -
Fast Cooling
1

15 kJ/in.
0.8 25 kJ/in.
40 kJ/in.
0.6
Category II -
Slow Cooling
0.4

0.2

0
0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
Wall thickness, in.

Figure 9. Limiting Methane Flow Rate for Procedures Qualified Using Flowing Motor Oil

100

90

80

70
Heat sink capacity, sec.

60
Category II -
15 kJ/in.
Slow Cooling
50 25 kJ/in.
40 kJ/in.
40

30
Category I -
Fast Cooling
20

10

0
0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
Wall thickness, in.

Figure 10. Limiting Heat Sink Capacity for Procedures Qualified Using Flowing Motor Oil

Copyright © 2002 by ASME 1515 Copyright © 2002 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/pdfaccess.ashx?url=/data/conferences/ipc2002/69711/ on 02/23/2017 Terms of Use: http://www.asme.org/ab