Technical Presentations at the April 2004 Meeting
1.1
Assessment
of Corrosion and Cracking Susceptibility of Welds,
Alan Turnbull (NPL)
An overview was given of work at NPL over the last few years in relation to the development of new or improved test methods for assessing the resistance of welds to corrosion and stress corrosion cracking. A starting point for that work was a major review of corrosion testing of welds undertaken in 1999 but available for general release only recently. A copy can be downloaded from:
The three major areas investigated included
detection of alloy depletion associated with sigma phase formation in a
super-duplex stainless steel (SDSS); preferential corrosion of carbon
steel welds, and a more recent investigation looking at factors affecting
stress corrosion testing of super 13 Cr steel welds.
SDSS (Corrosion, 58, 2002, 1039).
The primary objective was
to develop a method that would be sensitive to the Cr dealloyed zone
associated with formation of FeCrMo intermetallic and would indicate
whether sustained propagation was likely. It also had to be simple to use,
including field application. The critical pitting temperature test (CPT)
did not satisfy the latter (though examined for comparison) but also could
not distinguish welds with different heat input or indeed a failed weld.
Essentially, this is because the method responds to localised attack from
any source. Also, the CPT can be determined by a single active site and
thus does not reflect the extent of overall activity. In principle, the
electropotentiodynamic reactivation ( EPR) technique should be better and
indeed was more effective but also struggled because the region of sigma
phase formation was highly localised in the HAZ. Accordingly, using a
probe with 1 cm diameter exposed area, the effect of the sigma phase
region on the charge ratio was small and discrimination between different
welds was difficult. In principle, smaller probes could be developed but
it would be necessary to step along the surface and make a series of
measurements. The galvanostatic method developed at NPL simply involves
application of a constant anodic current (100 mAcm-2)
and monitoring the potential for one hour. If there is no significant
depletion the potential increases to the region of oxidation of chloride
ions or hydroxide ions and stays nearly constant. If there is a sensitised
region, the potential rises initially but following initiation of
localised attack, the potential decreases, often with noise indicating new
pits activating. The sustained fall, rather than recovery, is indicative
that pits would continue to propagate. The method worked well for the
controlled welds and was able to demonstrate sensitivity of the failed
weld from service.
Carbon steel welds (Br.
Corros. J., 37, 2002, 182)
Here, the challenge was
similar but focused on preferential corrosion of the weld or HAZ in
seawater. The galvanostatic method was adapted for this, the difference
being that visual examination was the only requirement. The principle is
elementary. In response to the application of an anodic current, the
material will yield the most susceptible regions preferentially. Visual
inspection is then used to detect regions of local mass loss. Depending on
the current the test could last from a few hours to a couple of days. Too
aggressive a test could lead to such general corrosion that discrimination
is not possible but 88 mA cm-2 for 2 hours was adequate to
reproduce a service weld failure.
Super 13 Cr martensitic stainless steel welds.
This work is ongoing and
involves testing of welds in internal fluid associated with oil and gas
pipeline as well as cathodically protected steel. The key issue is how
best to test and, in the latter, what constitutes a satisfactory
qualification criterion. We have undertaken miniaturised tensile testing
of samples taken from the weld metal, HAZ and parent plate at temperatures
up to 130 °C.
The data show a marked decrease in yield strength of the SDSS filler with
temperature, which raises the possibility (allied with creep) of
significant stress relaxation during 4 pt bend testing. Further work is
ongoing using FEA with these input tensile data and strain gauge specimens
to assess just how significant this is.
A parallel issue is the variability in through-thickness properties of a weld. Using a novel scanning microhardness technique, it is possible to readily discriminate local hardness regions; for example,e in the base of the weld root at the second pass interface, as shown below. Note that the black spots here are just artefacts from converting from the colour picture. The implication of such hardness variability arises when testing specimens machined from the weld that are thinner than full-wall thickness; the specimens may then not be representative. Interestingly, the hard zones corresponded to regions of compressive stress. It raises the more general question of the uncertainty in hardness in welds when tensile or compressive residual stresses are present.
Microhardness scan of super 13 Cr weld with super-duplex filler
1.2 Internet-based
Corrosion Information Systems,
Bob Cottis (Corrosion and Protection Centre, UMIST)
Corrosion information on the
web can be defined in general categories of general corrosion information
sites, electronic journals, scientific databases, corrosion education and
mailing lists.. Of the
corrosion information sites, Corrosion Doctors and Corrosion Source have
been found to be the most useful.
The www.jcse.org site is a site
to which scientific papers can be submitted for live refereeing.
Web sites of organisations such as NACE and CDA (www.copper.org) contain information on corrosion, while www.csa.com and www.cas.org run scientific abstract services, although the latter two require a subscription payment. On-line courses for corrosion engineering qualifications are run in the UK by www.umist.ac.uk which can lead to an MSc in Corrosion Control Engineering.
General
Web Pages
Corrosion
Doctors
CorrosionSource
Corrosion
Information Server
Corrosion.com
NACE
ICorr
Electronic
Journals
Open/free
Commercial/paid
for
Databases
Cor-Sur
(www.knovel.com)
Copper
& Copper Alloy Corrosion Resistance Database (www.copper.org)
Corrosion
Abstracts (www.csa.com)
Metadex
(www.csa.com)
SciFinder
(online version of Chemical Abstracts www.cas.org/SCIFINDER/scicover2.html)
Courses
Corrosion
Doctors
UMIST
Online Courses
Mailing
Lists and News
NACE
Corrosion Network
CORROS-L
Standardization
Activities
OCEAN
Open Corrosion Expertise Access Network
MatML
3.1 Control of Cathodic Protection on CRA Pipelines, Robin Jacob (The Corrosion Consultancy Limited)
Conventionally
designed sacrificial anode systems produce potentials more negative than
those strictly necessary to protect steels from seawater corrosion.
On pipelines, which are invariably coated, potentials will be close
to anode potential [ca. -1080 mV1].
Such negative potentials can have a number of undesirable effects,
including hydrogen embrittlement of some engineering materials
and premature deterioration of coatings.
As some pipelines are now being constructed from corrosion
resistant alloys, needed to handle corrosive internal conditions,
consideration needs to be given to the possibility of hydrogen
embrittlement.
The
presentation describes a method of controlling the potentials produced by
sacrificial anodes to values closer to those strictly required for
protection. By the use of
Schottky barrier rectifiers [SBR] in conjunction with the anodes, together
with careful design, potentials in the range -775 to -825 mV can be
ensured. At these
potentials, the risk of embrittlement is much reduced.
In order
to establish a design methodology for such limited potential systems, a
trial has recently been carried out using different types of SBR in a
number of configurations. Sufficient information is now available to allow
the design of limited potential cathodic protection systems operating
within a closely controlled potential range.
Data from a CRA pipeline fitted with potential control will be
presented.
[1
All potentials stated relative to
the Ag|AgCl|seawater reference electrode]
3.2 Keynote
lecture: Three New DNV Recommended
Practices on Submarine Pipeline Corrosion Control, Tomas Sydberger (DNV, Hoevik, Norway)
The following documents
were addressed:
- DNV RP-F102 (2003): Pipeline Field Joint Coating and Field Repair of Pipeline Coating
- DNV RP-F103 (2003): Cathodic Protection of Submarine Pipelines by Galvanic Anodes
- DNV RP-F106 (2003): Factory Applied External Pipeline Coatings for Corrosion Control
The
presentation briefly described the contents and philosophy of the three
documents and how they relate to other DNV documents on submarine
pipelines, as well as other national and international codes and
standards. Emphasis was made on DNV RP-F103 for which cathodic
protection (CP) design calculations have been performed for an oil export
pipeline with asphalt enamel + concrete coating and, furthermore, a
production flowline with a thermally insulating coating based on fusion
bonded epoxy and polypropylene. The
results were compared with those using other CP design codes, including
DNV RP B401, NORSOK M-503 (1997) and ISO/FDIS 15589-2 (2004).