Subscribe to our newsletter to receive the latest news and events from TWI:

Subscribe >
Skip to content

Submerged arc welding consumables. Part 3 - AWS specifications

   

Job Knowledge 89

Part 1
Part 2

As with the BS EN specifications for submerged arc welding consumables, the American Welding Society (AWS) system also uses a dual flux type/wire composition designation to identify the flux/wire combination that will provide the required properties.

The AWS system is somewhat simpler than the BS EN method, particularly if the full flux descriptor is used as specified in BS EN 760 (see Connect article No. 88). There are, however, only two specifications that deal with both wire composition and the flux but an additional two specifications that cover bare wires for stainless steels and the nickel based alloys. These are ANSI/AWS A5.17 - Carbon Steel Electrodes and Fluxes and ANSI/AWS A5.23 Low Alloy Steel Electrodes and Fluxes. The bare wire specifications are ANSI/AWS A5.9 Bare Stainless Steel Welding Electrodes and Rods and ANSI/AWS A5.11/A5.11M Nickel and Nickel Alloy Bare Welding Electrodes and Rods.

In AWS A.5.17 and AWS A5.23 the first part of the designation describes the flux type and may comprise up to six digits depending upon whether the flux is supplied with the tensile strength expressed in increments of 10 megapascals (two numbers where 43 represents 430MPa) or in pounds per square inch (1 digit ie 6 represents 60,000psi).

The first digit, the letter 'F', identifies the consumable as a submerged arc welding flux, the next letter 'S' is only included if the flux is made from or includes crushed slag. Omission of this letter 'S' indicates that the flux is unused and contains no crushed used flux introduced either by the flux manufacturer or the welding fabricator.

The next one or two digits specify the minimum tensile strength as explained above and this is followed by 'A' or 'P' for whether the test results were obtained in the as-welded, the A condition or post-weld heat treated, the P condition. The last digit identifies the minimum temperature at which a Charpy-V impact value of 27J can be achieved as in Table 1 below.

Table 1 Impact Test Requirements

DigitTest TemperatureImpact value
Joules
°C°F
Z no impact requirements 27
0 -18 0 27
2 -29 -20 27
4 -40 -40 27
5 -46 -50 27
6 -51 -60 27
8 -62 -80 27

In AWS A5.17 there is a total of eleven wires, split into three groups of low, medium and high manganese. The first digit, 'E', identifies the consumable as a bare wire electrode. If supplemented by 'C' the wire is a composite (cored) electrode. The composition of the solid wire is obtained from an analysis of the wire. However, since the composition of a cored wire may be different from that of its weld deposit the composition must be determined from a low dilution weld deposit made using a specific, named flux.

The next letter, 'L', 'M' or 'H' indicates a low (0.6% max), medium (1.4% max) or high (2.2% max) manganese content. This is followed by one or two digits that give the nominal carbon content. An optional letter 'K' indicates a silicon killed steel. There are a final two or three optional digits identifying the diffusible hydrogen in ml/100gms weld metal, H16, H8 or H4.

A full designation for a carbon steel flux/wire combination could therefore be F6P5-EM12K-H8. This identifies this as being a solid wire with a nominal 0.12% carbon, 1% manganese and 0.1 to 0.35% silicon capable of achieving an ultimate tensile strength of 60 k.p.i. (415MPa), a Charpy-V impact strength of 27J at -50°F (-46°C) in the post weld heat treated condition.

The classification in AWS A5.23 is, of necessity, rather more complicated as this specification covers a wide range of low alloy steels, a total of thirty one solid wires and twenty nine composite wire weld metal compositions. Within the confines of this brief article it will not be possible to cover in full the entire classification of the wires.

The flux designation is almost identical to that of AWS A5.17, except that a four, five or six digit identifier may be used. Why this additional sixth digit? Because some of the electrodes in the specification are capable of providing tensile strengths above 100,000 psi - in these cases the designation may be, for example, F11, identifying the flux as providing 110 ksi (760MPa) minimum tensile strength.

The classification of the wire comprises two parts - the first that of the wire, solid wires being prefixed 'E' and composite wires 'EC', the second part specifies the composition of the weld deposit. In Table 1 of the specification only the solid wires are listed. The wire classification commences with 'E' to identify a bare wire, the next letter places the wire in a 'family' of wires. 'L' or 'M' identifies the wires as being alloyed with copper, 0.35% max; 'A'as containing molybdenum, 0.65% max; 'B' as the creep resisting steels containing chromium and molybdenum; 'Ni' for those wires containing nickel. 'F comprises the Ni-Mo or Cr-Ni-Mo wires; 'M' triple de-oxidised Ni-Mo wires; 'W' aNi-Cu wire and 'G' not specified.

jk89f1.jpg

This use of wires to this latter 'G' designation may lead to problems as quite large changes can be made to the composition to achieve the required mechanical properties - a good example of this is where the NACE requirements for sour service of 248BHN or 1% nickel maximum are required. To achieve the required tensile or impact strength the consumable manufacturer may increase the carbon or nickel contents above those used in the procedure qualification test and still supply to the same designation.

Table 2 in AWS A5.23 classifies both solid wire and composite wire/flux combinations by means of weld metal compositions but still using the identifying letters as for the solid wires described above. The prefix 'E' is, however, omitted thus a carbon/molybdenum deposit may be classified, for example, as A3, a Cr-Mo deposit as B4, Ni-Mo as F5 etc.

Thus a full designation for a flux/wire combination for an as welded 1% Ni/0.25% Mo weld deposit with an ultimate tensile strength of 80ksi and an impact strength of 27J at -60°F (-51°C) may therefore be F8A6-ENi1-Ni1 and for a similar deposit using a cored wire in the PWHT'd condition F8P6-ECNi1-Ni1.

As mentioned in earlier articles on the topic of consumable specifications, it must be remembered that the mechanical properties and compositions are determined from test pieces taken from absolutely minimal dilution welds made on specified parent plates with a standard set of welding parameters - heat input, preheat, interpass temperature, post weld heat treatment temperature and time. They may therefore NOT reflect the results obtained in a production weld and the designation cannot be relied upon to guarantee the properties required by the application.

Where these properties are important it is therefore essential that mechanical testing, chemical analysis etc are determined from test specimens made using parent materials and parameters representative of production welding.

This article was written by Gene Mathers.