Validating the weldability with flux core arc welding “ FCAW ” for deteriorated naval steels ( FLUX CORED ARC WELDING )

Repairs and changes of steel plates in the structure of the vessels where it is possible to find steels with a high degree of deterioration need reliable welded joints in naval steel of the type ASTM A–131 Gr. A new to old. Due to the fact that the variables associated to the weldability of the materials to be repaired are not known, it is necessary to make a study on the weldability of steel ASTM A131 Gr A, in full penetration seams or junctures, under the various parameters involved in the welding process, to determine the influence of the corroding residues that affect the application of welding compounds through the analysis of the metallurgical reactions of liquid condition in order to select the contributing materials with the alloy elements that are able to prevent these phenomena and to recommend the best practices for the electric arc welding process with tubular electrode with FCAW flux core.

Studies on the weldability of ASTM A-131grams, naval steel with a high degree of deterioration are scarce.The objective of this study is to establish the behavior of these steels when they are subject to the FCAW (Flux Cored Arc Welding) processes.This process stands out for its high productivity and welding reduction costs as compared to manual processes.It is based on the fact that FCAW is a semi automatic process that reduces the periods of using welding, welding costs and also, they increase productivity by improving several indicators as compared to the SMAW process.
In order to validate the weldability, mechanic practices, micro structural behavior studies as well as studies on the behavior of several commercial tubular wires in Colombia will be conducted.
Welding with an electric arc with tubular electrode with flux core FCAW (Flux Cored Arc Welding) is a semi automatic welding process that takes advantage of an electric arc between a continuous electrode of filler metal and the welding electrode and the welding puddle.This process is used with the protection from a flux that can be found into the tubular electrode, with or without an additional gaseous protection from the outside, and without applying any pressure to it.Patarroyo, Solano, Cueca, Rojas, Morales Three stages were followed here: characterization of the naval steel, selection of the filler materials and the assessment of the weldability.

Experimental Design
Creating the parameters of the welding equipment was performed in order to qualify positions 2G and 3G correspondingly as per code number AWS D1.1:2004 The seam junctures were made with base materials according to the production years and service condition of the vessel.The nomenclature for the base material was set as follows: new material (N), old material (V) and old-welded material (VS); and the resulting combinations were N-VS, V-V and N-V for each one of the junctures part of the research.
The design of groove butt joints with root size face equivalent to 0 a 5 / 32 " root opening equivalent to 5 / 64 " a 1 / 16 " and bevel angle and chamfer of 45°.
The result analysis was done by simple comparison with: 18 observations in the tension test which corresponded to the 3 combinations per each one of the designation and 2 test tubes per coupon as per the Regulation, 36 observations in the side 3 combinations and 4 test tubes per coupon according to code AWS D1.1 and in the hardness test 36 observations for a total of 720 points obtained, 20 points per test tube).The metallographic analysis had 27 metallographies (3 for each combination and designation), for the base material, welding jobs and the area that had been affected by heat, done in the old material (V) and old welding jobs (VS) whose interest is to verify the weldability behavior in them for the FCAW process.The AWS D1.1 code criteria were applied here, obtaining the design of the joint of Fig. 2 for which the flame-cutting process was performed in order to generate the coupons on which the welded joints were done following the FCAW process with three electrode wires -1,-8 and -11.

Heat input H
Where:     As compared with the micro structure of naval steel from 2008, that shows an evenly distributed beaded ferritic structure as can be seen in Fig. 10.
The average Vickers analysis of hardness of the samples of the naval steel from 1969, 1972 and 2008, is consistent with the metallographic analysis.It was possible to obtain hardness indexes greater than that of the steel from 2008, supposedly due to the aging phenomenon due to the precipitation and migration of carbon on ferrite.
By applying the parameters in Table 1, and the design of the joint in Fig. 2, the FCAW welding process was applied with input -1, -8 and -11, according to what was stated during the experimental design the Joints welded using the visual inspection techniques, colored penetrating fluids, tension destructive test, side crease, metallography and hardness profile.
Regarding the side crease tests, the test tubes with the various combinations had a good behavior and the behavior is acceptable under the criteria in Code AWS D1.1.Characterization of a Welded Joint  Micrograph of a naval steel of the Type ASTM A-131 Grade A, with perlite (dark) dual phase on a ferritic matrix, Attacked with Nital at 2 %, 200 x enlargement.A ning perlite grain trend is evidenced here as well as some isolated hydrocarbons.
Micrograph of a naval steel of the Type ASTM A-131 Grade A, with perlite (dark) dual phase on a ferritic matrix, attacked with Nital at 2 %, 200 x enlargement.e presence of some isolated hydrocarbons is evidenced here.
Micrograph of a naval steel of the Type ASTM A-131 Grade A, with perlite (dark) dual phase on a ferritic matrix, Attacked with Nital at 2 %, 200 x enlargement.A ning perlite grain trend is evidenced here, some isolated hydrocarbons in the perlite grains limits.Micrograph of a naval steel of the Type ASTM A-131 Grade A, with perlite (dark) dual phase on a ferritic matrix, Attacked with Nital at 2 %, 200 x enlargement.e presence of some hydrocarbons is evidenced here.

Micrograph of a naval steel of the
Micrograph of a naval steel of the Type ASTM A-131 Grade A, with perlite (dark) dual phase on a ferritic matrix, attacked with Nital at 2 %, 200 x enlargement.e presence of some hydrocarbons is evidenced here.
Micrograph of a naval steel of the Type ASTM A-131 Grade A, with perlite (dark) dual phase on a ferritic matrix, Attacked with Nital at 2 %, 500 x enlargement.e inclusion of slug together with some hydrocarbons and grouped oxides is evidenced here.Micrograph of a naval steel of the Type ASTM A-131 Grade A, with perlite (dark) dual phase on a ferritic matrix, Attacked with Nital at 2 %, 200 x enlargement.

Micrograph of a naval steel of the
Micrograph of a naval steel of the Type ASTM A-131 Grade A, with perlite (dark) dual phase on a ferritic matrix, attacked with Nital at 2 %, 500 x enlargement.Some oxides embedded in the grain limits of the pearlitic phase are evidenced here.
Micrograph of a naval steel of the Type ASTM A-131 Grade A, with perlite (dark) dual phase on a ferritic matrix, Attacked with Nital at 2 %, 200 x enlargement.A ning grain trend of the pearlitic phase, generalized porosity in the ferritic matrix and some isolated hydrocarbons in the limits of the perlite grain can be seen here.
Micrograph of a naval steel of the Type ASTM A-131Grade A, with perlite (dark) dual phase on a ferritic matrix, Nital at 2 % attack, 200 x enlargement.Polygonal ferrite with hydrocarbons included and the presence of pores is evidenced here.Micrograph of a naval steel of the Type ASTM A-131 Grade A, with perlite (dark) dual phase on a ferritic matrix, Attacked with Nital at 2 %, 200 x enlargement.
Micrograph of a naval steel of the Type ASTM A-131 Grade A, with perlite (dark) dual phase on a ferritic matrix, attacked with Nital at 2 %, 500 x enlargement.Some oxides embedded in the grain limits of the pearlitic phase are evidenced here.
Micrograph of a naval steel of the Type ASTM A-131 Grade A, with perlite (dark) dual phase on a ferritic matrix, Attacked with Nital at 2 %, 200 x enlargement.A ning grain trend of the pearlitic phase is evidenced here.
Micrograph of a naval steel of the Type ASTM A-131Grade A, with perlite (dark) dual phase on a ferritic matrix, Nital at 2 % attack, enlarged 200 X.It is possible to see acicular perlite as well as some isolated oxides in the grain limits.Micrograph of a naval steel of the Type ASTM A-131 Grade A, with perlite (dark) dual phase on a ferritic matrix, Attacked with Nital at 2 %, 200 x enlargement.e presence of some isolated hydrocarbons, a ferritic matrix and an acicular pearlitic matrix is evidenced.
Micrograph of a naval steel of the Type ASTM A-131 Grade A, with perlite (dark) dual phase on a ferritic matrix, attacked with Nital at 2 %, 200 x enlargement.e presence of some isolated hydrocarbons, a ferritic matrix and acicular perlite are evidenced here.
Micrograph of a naval steel of the Type ASTM A-131 Grade A, with perlite (dark) dual phase on a ferritic matrix, Attacked with Nital at 2 %, 200 x enlargement.A ning grain trend of tempered and partially transformed perlite and some isolated hydrocarbons can be seen here.
Micrograph of a naval steel of the Type ASTM A-131Grade A, with perlite (dark) dual phase on a ferritic matrix, Nital at 2 % attack, 200 x enlargement.A ning grain perlite trend and some isolated hydrocarbons are evidenced here.INPUT 11 STEEL 1972STEEL 1972 Source: research results.
The average Vickers hardness for all base materials, shows that for years 1969-(143,51) and 1972-(143,41), it is relatively similar and they are higher when compared to the year 2008-(134,45).This is due to the probable hardening caused by its micro component precipitation and the years the materials have been used in the vessel.
The base materials in this study exhibited a ferritic matrix with perlite as the second micro component.Mechanical tension tests showed maximum load values, maximum effort values, rupture load and rupture effort within the ranges set by regulation ASTM A131 Grade A for designation -11.Designations -1 and -8 are below the values stated by the regulation and especially -8 whose rupture occurred in the input material.

Analysis and Conclusions
During welding process FCAW, some welding procedure specifications were generated, as well as the qualification of the welders.The conclusions were that the input materials designated as -1 and -8, require highly trained welders, because managing the welding puddle and especially the slag fluency result in inclusions and lack of penetration in the root of the joint.

Fig. 4 .
Fig. 4. Cutting and tagging the test tubes for chemical and metallographic analysis of the base metals.

Fig. 7 .
Fig. 7. Performing the tension tests on the test specimens.
Heat input in J/centimieters A = amperes V = electrical potencials in volts Va = Advancing speed in inches / minutes A = current in amperes V = voltage in volts va = advancing speed in inches per minutes 60 Heat intake for position 3G • Root pass = 53.647,05J/m • Filler pass = 52.258,06J/m • Presentation pass = 49.241,37J/m The base material of the test tubes obtained from the plates was characterized during the research, to verify it with ASTM A131 Gr A for Naval Steel, according to what is shown in Fig. 3.Chemical analysis tests were performed by spectrometry thus generating the chemical composition of metallographic analysis was performed and finding bead like segregation bands in the naval steel from 1969 and 1972, (see Figs.8 and 9) (boxes 1 and 4) on a ferritic matrix (box 3), supposedly due to the aging process due to the precipitation and migration of carbon on ferrite.Also sphere-shaped oxides were found in the ferritic structure (box 2), probably generated by corroding environments in an aqueous or gaseous environment.

Table 2
, from where it is easy to conclude that the naval steels manufactured in1969, 1972 and 2008, are typical steels of designation ASTM A131 Gr A.

Table 2 .
Chemical composition of the base material naval steel.

Table 5 .
Parameters of the Vickers hardness test.
Calculating the input heat for inputs -1,-8 and -11 generates heat affected areas that are typical in clean steels like ASTM A 131 Gr A.The metallographic analysis concluded that the steels from1969, 1972 and 2008, eventhough the first two exhibit pearlitic segregations, this does not affect the welded joints, including the old welding jobs found.A NV welding process should be graded for maintenance cases.FCAW welding process with AWS E71T-11input material exhibits the best welding results.It developed clean full penetration joints.This behavior allows for unilateral welds in vessels, thus displacing the SMAW welding methods which have partial penetration and root clean up.ORTÚZAR, Raúl.Mecánica de fracturas en estructuras navales.En: Congreso Panamericano de Ingeniería Naval "XVI COPINAVAL 99"1988 CASTILLA, Iván J. UNFRIED, Jimmy S. Soldabilidad de un acero microaleado utilizando el proceso SMAW y metal de aporte ferrítico de alta resistencia.Revista Ciencia y Tecnología de Buques Numero 3 Volumen 2. Julio de 2008.