Submerged Arc Welding Fluxes:
The criticality of particle size

by Naddir M Patel
(Calgary, Alberta)

(First Part)
Background:

Submerged Arc Welding (SAW) is a very versatile process that can be used across a range of materials and plate thicknesses for fabrication of water and petrochemical pipelines, gas cylinders, ship building and repair, and resurfacing (hardfacing) applications in the mining, mineral processing and power industries.

One such application is the high speed welding (at 450-500A; 28-32V; 1400-1500mm/minute, 2 weld runs) of LPG (liquid petroleum gas) cylinders in India.

About 10-12 years back, while I was managing Silico Products, a company manufacturing Submerged Arc Welding fluxes in Mumbai, India, we were approached by some five of our best customers, who were manufacturing LPG (liquid petroleum gas) cylinders, with a complaint that their current SAW operation resulted in up to a 20% weld integrity reject rate. It was clearly an unsustainable position.

Electrode specifications in terms of chemistry and wire diameter are very clear: under the classification F65A0-EL8, of AWS A5.17 (Specification for Carbon Steel Electrodes and Fluxes for Submerged Arc Welding), the electrode is well defined (EL8 is an AWS D1.1 designation for a Cu-coated mild steel wire with nominal composition: C = 0.1% max, Mn = 0.4-0.6%, Si = 0.03% max, Cu = 0.14% max., P and S = 0.03% max.), as are the mechanical properties of the expected weld metal.

No details however are forthcoming about the other consumable, the SA welding flux, which is a critical component for achieving a sound weld that, due to its proprietary nature, has always been restricted to a gray area, with the consumer having to depend almost totally on the vendor, and chalking off defects to unavoidable processing conditions.

The Mild Steel plate thickness being barely 3mm in the above application (HRC hot rolled coil of thickness 2.9mm and nominal composition: Mn = 0.9%, Si = 0.25%, C = 0.2%, P and S = 0.035% max.), the problem of burn-through and weld defects involving oxide inclusions and porosity was a serious time and cost multiplier.

As process parameters required 2 runs, (circumferential and bung) the challenge was, therefore, to design a uniform melting and fast self-peeling flux able to handle welding abuse (unclean surfaces, no pre-heat, no flux pre-heat etc.) which is very endemic within the unregulated welding sectors of India.

As the major suppliers of SAW flux to a wide range of industries, we could not remain indifferent to those grievances. We had to start doing something immediately.

It must be pointed out that Submerged Arc Welding fluxes, are complex, multipurpose ceramic compositions whose mineral and alloy ingredients are mixed together in proprietary combinations, and processed (either by electric fusion or agglomeration) to yield a granular ceramic product.

These granulated products are used as consumables, in various proportions, with the welding wire in Submerged Arc Welding applications. The flux, physically deposited to cover the welding arc zone (hence the name Submerged Arc) shields the weld arc zone from the atmospheric contamination.

Unlike the exclusively shielding characteristics of gas combinations used in the MIG/TIG processes, the SAW flux takes on the triple role of shielding the weld from the atmosphere, refining the weld metal (addition of alloying elements and removal of tramp oxides), and peeling off as a slag once the welding is complete.

Like any chemical reaction, the flux has to be designed (formulated) to incorporate Thermodynamics, Kinetics and Transport phenomena. It must:

1. Melt just below the temperature of the steel being welded via Material balance and phase diagrams to achieve the ideal eutectic point. (Thermodynamics).
   
2. Mix with the parent material in the molten zone and refine the weld metal, adding elements such as Mn, Si, Cr, etc and removing rust and non metallic oxide inclusions from the weld zone, by enveloping the oxides in a Silicate-aluminate matrix. (Kinetics).
   
3. Float up (Transport the oxides) to the surface before the steel solidifies and peel off automatically (self lifting slag).

Thus the formulation must melt uniformly at specific temperatures and possess operational characteristics such as specific gravity and fluidity to refine the weld metal and surface tension characteristics to ensure speedy slag peel off.

An action plan was formulated to collect process data and to research the welding process. We set forth a standard report to follow on current production, and asked the manufacturers to report details of their operations.

The main data we sought were Operational Parameters (A, V, welding speed, parent material thickness and chemistry and electrode chemistry) and defect identification (burn through, porosity, slag inclusions, pock marks, surface discoloration etc.).

Whereas the operational parameters were immediately available, there was resistance to the supply of weld defect identification and tabulation documentation. The management was concerned about information leaking out and the union was worried that this could be a tool used for decreasing productivity based wages.

A compromise was arrived at, in that we supplied the manpower with a very specific mandate not to identify the weld station or the welder and to collect weld defect data strictly on a cylinder/weld run basis. We of course guaranteed that the data would stay strictly confidential with us and there would be no identification of the manufacturer.

Utilizing a VOC (Voice of Customer) methodology, an operational data collection drive was set-up to monitor cylinder welding. Defects were then stratified and plotted as Pareto charts. Fish bone (Ishikawa) diagrams were then set up to arrive at the root causes of the defects.

Based on a sampling number confidence level of 95%, weld defect data were then collected over a week (6 days) evenly divided between 2 shifts and 4 welding stations.

(to be continued in a new page)