W ELDING OF PIPELINES AND RELATED FACILITIES 17 API 1104 – 2010 EDITION

If any of the following essential variables are changed in a

procedure specification, the welder using the new procedure
shall be requalified:
a. A change from one welding process to another welding
process or combination of processes, as follows:
1. A change from one welding process to a different
welding process; or
2. A change in the combination of welding processes,
unless the welder has qualified on separate qualification
tests, each using the same welding process that is used for
the combination of welding processes.
b. A change in the direction of welding from vertical uphill
to vertical downhill, or vice versa.
c. A change of filler-metal classification from Group 1 or 2
to Group 3 or from Group 3 to Group 1 or 2 (see Table 1).
6.4 VISUAL EXAMINATION
For a qualification test weld to meet the requirements for
visual examination, the weld shall be free from cracks, inadequate
penetration, and burn-through, and must present a neat
workman-like appearance. The depth of undercutting adjacent
to the final bead on the outside of the pipe shall not be
more than 1/32 in. (0.8 mm) or 12.5% of the pipe wall thickness,
whichever is smaller, and there shall not be more than 2 in.
(50 mm) of undercutting in any continuous 12 in. (300 mm)
length of weld.
When semi-automatic or mechanized welding is used,
filler wire protruding into the inside of the pipe shall be kept
to a minimum.
Failure to meet the requirements of this subsection shall be
adequate cause to eliminate additional testing.
6.5 DESTRUCTIVE TESTING
6.5.1 Sampling of Test Butt Welds
To test butt welds, samples shall be cut from each test
weld. Figure 12 shows the locations from which the specimens
are to be removed if the test weld is a complete circumferential
weld. If the test weld consists of segments of pipe
nipples, an approximately equal number of specimens shall
be removed from each segment. The total number of specimens
and the tests to which each shall be submitted are
shown in Table 3. The specimens shall be air cooled to ambient
temperature prior to testing. For pipe with an outside
diameter less than or equal to 1.315 in. (33.4 mm), one fullpipe
section specimen may be substituted for the root-bend
and Nick-break specimens. This full-section specimen shall
be tested in accordance with 5.6.2.2 and shall meet the
requirements of 6.5.3.
6.5.2 Tensile-strength, Nick-break, and Bend-test
Procedures for Butt Welds
The specimens shall be prepared for tensile-strength, Nickbreak
and bend tests, and the tests shall be performed as
described in 5.6. However, for the purpose of welder qualification,
it is not necessary to calculate the tensile strength of
the coupons. The tensile strength test may even be omitted, in
which case the specimens designated for the test shall be subjected
to the Nick-break test.
6.5.3 Tensile-strength Test Requirements for
Butt Welds
For the tensile-strength test, if any of the reduced-section
specimens or the full-section specimen breaks in the weld or
at the junction of the weld and the parent material and fails to
meet the soundness requirements of 5.6.3.3, the welder shall
be disqualified.
6.5.4 Nick-break Test Requirements for Butt Welds
For the Nick-break test, if any specimen shows imperfections
that exceed those allowed by 5.6.3.3, the welder shall be
disqualified.
6.5.5 Bend Test Requirements for Butt Welds
For the bend tests, if any specimen shows imperfections
that exceed those allowed by 5.6.4.3 or 5.6.5.3, the welder
shall be disqualified. Welds in high-test pipe may not bend to
the full U shape. These welds shall be considered acceptable
if the specimens that crack are broken apart and their exposed
surfaces meet the requirements of 5.6.3.3.
If one of the bend test specimens fails to meet these
requirements and, in the company’s opinion, the imperfection
observed is not representative of the weld, the test specimen
may be replaced by an additional specimen cut adjacent to the
one that failed. The welder shall be disqualified if the additional
specimen also shows imperfections that exceed the
specified limits.
6.5.6 Sampling of Test Fillet Welds
To test fillet welds, specimens shall be cut from each test
weld. Figure 10 shows the locations from which the specimens
are to be removed if the test weld is a complete circumferential
weld. If the test weld consists of segments of pipe
nipples, an approximately equal number of specimens shall
be removed from each segment.

9.7 VISUAL ACCEPTANCE STANDARDS FOR

UNDERCUTTING
9.7.1 General
Undercutting is defined in 9.3.11. The acceptance standards
in 9.7.2 supplement but do not replace visual inspection
requirements found elsewhere in this standard.
9.7.2 Acceptance Standards
When visual and mechanical means are used to determine
depth, undercutting adjacent to the cover or root bead shall
not exceed the dimensions given in Table 4. When both
mechanical and radiographic measurements are available, the
mechanical measurements shall govern.