Submerged Arc Welding (SAW): Definition, Process & Uses

A shipyard welder once joked that submerged arc welding is the only process where you never actually see the arc you’re being paid to run — the light, the spatter and most of the heat simply vanish under a gray mound of granular flux. That quiet mound is exactly why a process invented in the 1930s still welds the thickest, longest, most load-critical joints on the planet.

What Is Submerged Arc Welding? A Clear Definition

The most direct way to define submerged arc welding (SAW) is this: it is a fusion welding process that forms coalescence between metals by heating them with an arc between a bare metal electrode, or multiple electrodes, and the workpiece. This arc is completely buried underneath a blanket of granular fusible flux, which is why the process earns its name of being “submerged”.

In line with the American Welding Society’s definition, SAW produces welds with no applied pressure, and the filler metal is supplied by the electrode itself, or at times by a supplementary strip or rod. As noted in practical AWS guidance, this process uses that flux blanket to shield the arc and molten weld pool, meaning no external shielding gas is used whatsoever. The result is an arc that cannot be seen by the naked eye, with virtually no visible arc light, and far less spatter than most open-arc processes.

This shielding works on two fronts. First, as the granular flux melts from the heat of the arc, it decomposes and creates a protective vapor dome around the pool. Second, the remainder forms a molten slag that floats over the weld metal, protecting it as it cools and solidifies. Once cooled, much of that slag will break away on its own, leaving behind a smooth, uniform bead. For reference, SAW falls under the AWS A5.1/A5.17 family of filler metal and flux specifications, as well as ISO 14191 for process classification.

The Basic Operating Picture

The setup for SAW is built around continuity. A continuously fed, bare wire electrode is driven by a wire feeder down through a welding head. Directly ahead of the arc, a flux hopper dispenses a controlled mound of granular fusible flux over the joint. The arc is struck between the end of that wire and the workpiece, instantly submerging itself under the flux. From there, the head travels along the joint, melting the wire, the base metal, and portions of the flux to form the weld pool and slag. Unused flux can then be reclaimed by a recovery system to be reused when viable.

Common Arrangements

While a single-wire setup is the most basic form of SAW, there are a few arrangements used to tailor the weld to the job:

• Primary (single-wire): One continuously fed electrode. This is the simplest configuration and is used for a wide range of straight, long joints.
• Twin-wire: Two electrodes fed side by side. This increases deposition rate while still maintaining a relatively stable, controlled bead.
• Series arrangement: Two electrodes placed in line along the direction of travel, sharing the same circuit. This concentrates heat to achieve deeper penetration for thicker joints.
• Broadside arrangement: Multiple wires spread across the width of a joint or cladding area, allowing for wider welds in a single pass.

All of these arrangements are almost exclusively mounted to mechanized equipment, such as a welding tractor, gantry, or column-and-arm manipulator. At its core, regardless of how many wires are added, the defining trait never changes: the arc is always buried under granular flux, with no external shielding gas, and the process relies on fusion to create a coalesced joint. That single trait is what separates SAW from nearly every other arc welding process.

How the SAW Process Works, Step by Step

Watch a submerged arc welding (SAW) rig from a few feet away and it looks almost boring: a tractor crawls along a seam, wire feeds into a gray pile, and nothing dramatic happens. Underneath that pile, though, a fully shielded fusion process is running at hundreds of amps. Here is the operating sequence, in plain language, from first clean to finished bead.

Step 1: Joint Preparation and Fit-Up

Everything starts with clean steel. The joint faces are beveled, the plate is aligned to the drawing, and mill scale, rust, oil and cutting residue are removed from at least an inch either side of the seam. Surface preparation matters more in SAW than most people expect: because the arc is buried, the welder cannot see contamination being burned out, and the flux will happily carry rust or oil straight into the pool. Joints are tacked, run-on and run-off tabs are attached where full-length bead integrity matters, and the assembly is checked for the root gap and root face specified in the procedure.

Step 2: Flux Delivery and Arc Initiation

A hopper above the welding head gravity-feeds granular flux through a dispensing nozzle, laying a mound that buries the wire tip completely. The AWS practical guide to SAW describes the standard blanket as roughly 25 to 35 mm deep over the arc zone. Once the flux is in place, the arc is struck either by a retract start (the wire pulls back, touches, and lifts) or with a high-frequency assist. Within a second the wire, the plate and the flux are all melting together.

Step 3: The Arc, the Chemistry and the Shielding Dome

This is where the process earns its name. The arc melts the bare wire electrode, the base metal underneath it, and the fusible flux immediately around it. As described in ScienceDirect’s overview of the process, that flux does three jobs at once:

• It decomposes into a shielding gas dome that isolates the molten pool from oxygen and nitrogen, so no external shielding gas cylinder is needed.
• It deoxidizes and refines the pool, with elements like manganese and silicon scavenging oxygen before it can form blowholes.
• It forms a molten slag that floats on the bead and protects the metal as it freezes.

The result is a quiet, spatter-free weld with an intense but invisible light source underneath a blanket of unreacted granules.

Step 4: Solidification, Slag Removal and Flux Recovery

As the head moves on, the pool solidifies under the slag layer. A correctly formulated flux produces slag that lifts off in one continuous piece after cooling, often curling up on its own, leaving a smooth as-welded surface that rarely needs grinding. Any flux that did not melt is picked up with a vacuum recovery hose, screened, and returned to the hopper for reuse, which is one of the quiet economic advantages of the process.

Step 5: Flux Depth and Granulometry, the Variables That Bite

Flux depth is the single most touched adjustment on a SAW head, and both extremes cause trouble:

• Too shallow: the arc blows out of the mound. You get visible arc flash, ultraviolet radiation, spatter, and the fume and light exposure the process is designed to avoid.
• Too deep: trapped gases and un-melted granules cannot escape cleanly, producing porosity, slag entrapment, and lack of fusion at the toes.

Granulometry matters equally. Coarse fractions flow and recover well but shield less consistently; very fine powder packs too tightly, vents poorly, and raises porosity risk. Most flux suppliers publish a usable particle band, and blending fractions outside it is a common shortcut that shows up later as weld rejects. The process breakdown from Xometry sums it up neatly: flux condition, not arc visibility, is the real process indicator.

Once those five steps are stable, the rest of the process, the equipment, parameters and joint design, is really about holding this cycle in balance from the first inch of the bead to the last.

SAW Equipment and Consumables: What You Need

One useful way to picture a submerged arc welding station is as a small, self-contained fabrication line on wheels. Nothing is hand-held. The power source, wire feeder, flux hopper, tracking tractor and recovery gear move together as one unit, laying a bead that can run for meters without stopping. Understanding what each piece does makes the rest of the process much easier to follow, because in SAW the equipment carries most of the skill that a welder normally carries in their hands.

The Core System, Piece by Piece

At the heart of the setup sits the power source. Most modern SAW machines are constant-voltage rectifier or inverter types, although generator-type machines still show up on pipeline and field work. SAW runs happily on either DC or AC; a practical overview of the typical configurations is covered in the AWS Welding Digest practical guide to submerged arc welding.

Next comes the wire feeder, which pushes bare electrode from a spool at a speed matched to the power setting. Because SAW normally runs on constant voltage, the arc is largely self-regulating: if the arc shortens, current rises, deposition increases and the arc pushes back to length. That stability is a big reason SAW automates so cleanly.

Mounted above the arc is the flux hopper and dispensing nozzle, a small hopper that gravity-feeds granular flux through a tube directly in front of the electrode. The wire feeds through a welding head centered in that stream of granules. Behind the head, a flux recovery unit — essentially an industrial vacuum with a suction hose — lifts unused granules off the finished bead and returns them to the hopper. Only flux that has actually melted and fused to the bead is consumed, so recovery keeps costs down and waste manageable.

Movement comes from a welding tractor (a compact, track-mounted automator) for long straight seams, or from a column-and-arm manipulator or gantry when the workpiece is large and stationary, such as a pressure vessel shell or a wind tower section. The tractor simply rides a rail or the plate edge and drags the head at a set travel speed.

Wire Electrodes: Classes and Diameters

SAW consumables are standardized, which is what makes procedure qualification possible. Carbon steel bare wires are classified under AWS A5.17, with common designations such as EM12, EM12K and EH14. The letters encode chemistry and deoxidizer content: the “K” suffix indicates a wire with silicon and manganese deoxidizers suited to a wider range of plate conditions, while EH14 wires carry higher manganese for tougher deposits. Low-alloy steel wires fall under AWS A5.23, covering grades for chromium-molybdenum, nickel-bearing and high-strength applications — a step up in class that mirrors the way alloy work is handled with Alloy Steel Welding Electrodes: Selection, Use & Safety.

Wire diameters typically run from about 2.0 mm to 5.0 mm (roughly 5/64 to 3/16 inch). Thinner wire gives fine control on lighter beads; fatter wire pushes current density up and enables very high deposition rates. Beyond solid wire, shops can run metal-cored or flux-cored wires for faster fill and better tie-in, and wide strip electrodes used with electroslag techniques to clad stainless or nickel-based layers onto carbon steel plate.

Flux Types: Fused and Bonded

Flux is a consumable as real as the wire. Fused flux is pre-melted, granulated and dried, so it is dense, moisture-resistant and extremely consistent from batch to batch; it also recovers well through vacuum systems. Bonded flux is mixed and chemically bonded at lower temperature, costs less and can carry ingredients that fused processing would destroy, but it absorbs moisture from the air and is noticeably more sensitive to rust, oil and grease on the plate. Store bonded flux like a sponge is stored in a desert: sealed, dry and warmed before use. Incidentally, this whole flux chemistry world is steel-specific — aluminium takes a different route, as explained in Aluminum Flux Welding: How It Works and When to Use It.

Multi-Wire and Tandem Heads

Productivity scales by adding wires. A twin-wire head feeds two electrodes into one arc zone; series arrangements pair two wires on a single AC source so they share one arc, while double-DC setups give each wire its own power source. Tandem heads line two or more wires in sequence along the travel direction, letting the lead wire dig and the trailing wires fill, often with independent voltage control per wire. Broadside arrangements place wires side by side to widen the bead in a single pass. Each step up the ladder raises deposition rate without raising travel time, which is the economic engine behind high-volume SAW production and a key argument in the case to consider submerged arc welding to improve productivity.

Welding Parameters and Their Effect on Weld Quality

On paper, submerged arc welding looks almost hands-off: the arc disappears under flux, the tractor drives itself, and the operator mostly watches. In practice, the craft of SAW happens before the arc is ever struck. Almost every weld quality problem you will meet in heavy plate — a washed-out bead, a cold lap, a row of porosity, a brittle heat-affected zone — traces back to one of about seven settings. Learn what each dial actually controls and SAW becomes remarkably predictable.

The Seven Variables You Actually Adjust

Parameter	What it mainly controls
Welding current	Penetration depth, deposition rate, bead height
Arc voltage	Bead width, profile shape, how far the flux melts ahead of the arc
Travel speed	Heat input per millimetre, bead thickness, tie-in at the toes
Wire feed speed	Deposition volume; on constant-voltage machines it is the main current setter
Stick-out (electrical contact length)	Resistive heating of the wire, arc stability, effective current
Flux depth over the arc	Containment of the arc and shielding dome (covered in depth earlier)
Electrode angle and spacing	Direction of arc force, bead overlap in twin and tandem heads

Think of voltage and current as the two that shape the bead, and travel speed as the one that decides how much total energy the metal receives. Voltage stretches the arc and widens the weld pool, giving a flatter, broader profile; drop it too low and the bead stands proud with a narrow, high-strength root. Current does the heavy work: SAW runs at unusually high current density through a small contact area, which is exactly why the arc drives deep into the joint in a single pass rather than merely capping the surface. Wire feed speed and current are locked together on the constant-voltage power sources used in most SAW cells, so raising the feed rate is usually how you raise amps.

Heat Input: The Number Behind Toughness

Travel speed converts those settings into metallurgy. Welders calculate heat input as:

Heat input (kJ/mm) = (amps × volts × 60) ÷ (travel speed in mm/min × 1000)

That single figure governs cooling rate. Cool slowly from a high heat input and the heat-affected zone grains coarsen, which can cost you notch toughness; cool too quickly and you risk hard, hydrogen-sensitive structures in higher-strength steel. As a practical AWS guide to submerged arc welding puts it, parameter selection is really procedure control in disguise. The reward for getting it right is productivity: SAW deposition rates run roughly four to ten times those of manual stick welding, with figures cited up to about 45 kg/h on heavy-plate work, compared with the far slower, stop-start rhythm of Advantages of Stick Welding: Why SMAW Still Wins.

Stick-out deserves special mention. It is the length of wire energised between the contact tip and the arc, normally held around 25–35 mm. Within that band, lengthening the stick-out adds resistive heat in the wire and raises current and penetration; go well beyond it and the extension wanders, the arc becomes erratic, and penetration becomes inconsistent instead of deeper.

Bead Profile Defects Traced Back to Parameters

• High voltage, low current: a wide, thin bead with washed-out toes, shallow penetration and a real risk of lack of fusion at the joint walls.
• Excessive current: deep and fast, but it can burn through the root, leave undercut at the toes, or drive gas out of the pool fast enough to trap porosity.
• Travel speed too high for the settings: a narrow, high-shouldered bead that ties in poorly and concentrates heat in a thin band.
• Wrong stick-out: wandering arc, spatter under the flux, inconsistent penetration across the length of a seam.
• Poor multi-wire spacing or angle: overlapping beads, cold laps between wires, or an uneven crown on the final pass.

Once the procedure is qualified, the daily job is simple: hold voltage, current, speed and stick-out steady, keep flux depth consistent, and let the tractor do the rest. Consistency, not cleverness, is what makes a submerged arc weld look the same from the first metre to the last.

Joint Designs, Welding Positions and Techniques

Submerged arc welding is a picky process when it comes to geometry. It deposits metal faster than almost any other arc process, but it will only do it from one orientation: with the joint sitting flat. Understanding where SAW fits — and where it does not — is the difference between a shop that uses it to knock out heavy plate work in hours and a shop that keeps fighting sagging beads, lost flux and burned-through edges.

Why SAW Stays in Flat and Horizontal

SAW is normally restricted to the flat (1G) and horizontal (2G) positions, and the reason comes straight down to physics. The arc is buried under a mound of loose granular flux. In a flat butt joint the flux cradles the arc, the molten pool sits inside a self-contained mould of slag and unfused granules, and gravity works in your favour. Tip that joint up to a vertical or overhead position and the granules trickle off, the fluid pool sags, and the arc is no longer submerged at all. As AWS’s practical guide to the process notes, SAW is fundamentally a downhand, mechanized process, and shops routinely build fixtures, turning rolls and gantries around that constraint rather than trying to fight it.

Horizontal fillet welds (2F) are the practical exception, and they work because the root of the joint still supports the pool. Everything else — vertical, overhead, field positions, pipe in fixed 5G or 6G — is off the table. If a project has out-of-position joints and small lot sizes, SAW is the wrong tool and an all-position process is the right one; our All Position Welding Rod Guide covers what to reach for instead.

Common Joint Types for SAW

The joint designs that SAW handles well are all variations on plate butt welds and fillets:

– Square butt joints with a small root opening. This is SAW’s signature advantage. Because a single-wire submerged arc at normal settings concentrates heat deeply, a square edge joint with a root gap of a few millimetres can be welded full-penetration in one pass. That eliminates beveling entirely, which saves both machining time and a startling volume of filler metal on long seams.
– Single-V and double-V grooves. For thicker plate, bevels are still used, but SAW tolerates much narrower included angles than open-arc processes. Groove angles of roughly 45 degrees per side are common, and qualified multi-pass procedures often go shallower still, which again cuts deposited volume.
– T-joints and fillet welds. Lap and T joints in flat position weld cleanly, with deep throat build-up per pass and very consistent leg dimensions thanks to mechanized travel.
– Edge joints. Common in fabricated plate assemblies where only the edges need fusing.

Two supporting details matter in practice. The first is backing: full-penetration butt joints in heavy plate are usually welded with a steel backing strip, or with a flux-backed free-edge joint where a layer of granular flux is placed under the root opening so the penetrating weld metal is supported from below. Double-sided welds are the alternative: weld one side, clean and inspect, then back-chip the root to sound metal before welding the second side so no unfused or oxidised material is trapped in the joint.

The second is run-on and run-off tabs. Because arc starts and craters are the weak points of any weld, SAW procedures tack small steel tabs at both ends of the seam so the start and stop happen on disposable metal, outside the finished joint.

Single-Pass vs Multi-Pass Welding

Thin assemblies are usually a single-pass affair. Heavy sections — pressure vessel shells, wind tower sections, bridge box girders — are built up in layers. The typical approach is a root pass, a fill sequence of stringer beads laid side by side with controlled overlap, and a final cap pass with slightly higher voltage for a smooth profile. Interpass temperature is monitored, cleaning between layers is mandatory, and any flux or slag trapped between beads becomes a permanent inclusion, which is why operators vacuum the groove religiously between passes.

The practical takeaway: design the joint around the process. Give SAW square edges where penetration allows, qualify the groove angle down where volume is a problem, plan backing or back-chipping for full-penetration work, and never assume a position the flux cannot hold. Shops that do this consistently find that SAW joint prep is cheaper and simpler than for almost any competing heavy-plate process, which is exactly why it still dominates long seams in shipyards and vessel shops.

Advantages of Submerged Arc Welding

Ask a fabrication shop why it bought a submerged arc welding (SAW) system and the answer is usually speed. Speed, though, is only the headline over a set of advantages that compound each other: SAW lays down metal faster, finishes cleaner, consumes less filler per kilogram deposited, and asks far less of the operator standing beside it. That combination is why the process dominates long, heavy, repeatable welds in shipyards, pressure-vessel shops, pipe mills and structural fabrication. As Hobart Brothers’ technical guidance on SAW productivity puts it, the process was built for shops where deposition rate and consistency drive the cost of a weld more than hourly labor does.

Deposition Rates and Travel Speeds That Open-Arc Processes Can’t Match

The biggest economic lever in welding is how much sound metal you deposit per hour. SAW pushes that figure up because nothing is wasted: there is no electrode stub end, no shielding gas escaping into the atmosphere, and almost no spatter. A continuously fed bare wire running at high current can melt at rates that make stick welding look like a hobby.

Practical figures vary with wire diameter, polarity and alloy, but AWS’s practical guide to SAW in Welding Digest and vendor field data consistently put single-wire systems in the range of roughly 4–10 kg/h (about 9–22 lb/h), with twin-wire, series and tandem multi-wire heads pushing well past 20 kg/h on heavy butt joints. A mechanized tractor also travels continuously, so a 20 mm butt weld that would consume an hour of a stick welder’s arc time can often be completed in a fraction of that.

Deep Penetration With a Clean, Finished Bead

SAW’s high current density at the wire tip drives heat deep into the joint, which means fewer passes to fill a given groove. Fewer passes means less filler metal, less flux, fewer arc starts and stops, and less total heat input into the part.

The visible result is equally valuable. Because the arc is buried, there is no wind-disrupted gas envelope and no wander, so bead geometry stays consistent from the start of a run to the end. The bead comes out smooth, uniform and almost entirely free of spatter, and in applications where the face is not cosmetic, it needs little or no grinding or dressing afterward. That removes an entire finishing operation from the production flow.

Less Fume, Less Light, Less Operator Fatigue

Open-arc processes throw off intense ultraviolet light, spatter and a plume of fume that must be captured or ventilated around. Because SAW’s arc sits under a blanket of flux, most of that output never escapes. Compared with flux-cored or stick welding on the same joint, SAW generates far less visible light, far less spatter, and measurably less airborne fume per kilogram of deposited metal, a claim supported in academic overviews of the process.

For the operator this translates into real comfort. There is no arc flash to shield against while monitoring travel, no spatter burning holes in clothing, and much less eye strain, so operators often work without a hand-held shield while watching the flux mound and wire feed. Over an eight-hour shift that lowers fatigue, and lower fatigue correlates with fewer mistakes and fewer rework loops.

Automation, Repeatability and Reduced Skilled-Labor Dependence

SAW is almost never hand-welded, which sounds like a limitation until you consider what it buys. Once a procedure is qualified, the machine repeats it all day without drifting: bead width, penetration depth and travel speed stay inside the narrow band the procedure was qualified to, shift after shift. This is how a shop moves quality control out of the welder’s hands and into the written procedure and the parameter sheet.

Because the operator supervises a tractor or gantry rather than controlling a puddle by eye, SAW also reduces dependence on hard-to-find all-position specialists for the shop’s highest-volume work, and trainees reach useful productivity faster than on manual pipe or structural jobs — a pattern familiar to anyone who has seen how demanding all-position welding actually is when done by hand. Gantry rigs, column-and-arc manipulators and robotic cells all carry SAW heads, and shops welding alloy steels and heat-treated plate especially value the tight parameter control, since mechanical properties in those materials are sensitive to inconsistent heat input.

Lower Distortion and Better Mechanical Properties

Deep, concentrated penetration fills joints with fewer passes, and fewer passes mean fewer thermal cycles across the plate. The result is lower residual stress and less angular distortion per millimetre of weld metal deposited — an advantage that compounds on large assemblies, where distortion is the difference between a frame that fits and a frame that gets scrapped. SAW fluxes are also designed to deoxidize and refine the pool, so deposited metal typically shows a clean, sound structure with good impact and tensile results when filler and flux are matched under a qualified procedure.

Where the Advantage Really Shows Up

These benefits explain SAW’s economics: the process concentrates its value on long, straight, thick, high-volume welds — longitudinal seams on rolled plate, circumferential seams on vessels, girder flanges, tank floors and structural butt joints. On that work the cost per metre of deposited, code-compliant weld metal is typically lower than any manual or semi-automatic alternative. On short, out-of-position or one-off work the advantage largely disappears, and knowing where to draw that line is the real skill in justifying a SAW system.

Limitations, Common Defects and How to Fix Them

Submerged arc welding earns its reputation for clean, uniform, high-quality beads, but no process is flawless. Shops that switch to SAW expecting perfection usually discover a simple reality: the process is highly tolerant of parameter mistakes on a well-prepared joint, and highly intolerant of a dirty, poorly fitted one. Knowing where SAW struggles is just as valuable as knowing where it shines.

Inherent Limitations of the Process

The first limitation is gravitational. Because the molten pool is held in place by a mound of flux and a blanket of slag, SAW is essentially a flat and horizontal process. Nearly all production work happens in the 1G flat position or, less often, 2G horizontal. Out-of-position welding would let the liquid pool run out from under the flux and expose the arc. If a job requires vertical or overhead work, other processes are the answer, and it pays to understand the all position welding rod guide before assuming SAW can be adapted.

Other practical limitations include:

• Joint preparation demands. SAW likes clean, tight, well-aligned joints. Mill scale, rust, oil, moisture and wide root gaps all translate directly into defects.
• Flux handling. Flux must be stored dry, dispensed consistently, recovered, screened and either reused within limits or disposed of.
• Residual flux disposal. Unused recovered flux, especially flux that has picked up iron powder and moisture, cannot simply be dumped on the shop floor. It becomes a housekeeping and environmental cost.
• Thin material. The high current density that makes SAW productive also makes burn-through on thin sheet a constant risk, so light-gauge fabrication is generally a poor fit.
• Setup cost and floor space. Tractors, gantries, hoppers and recovery vacuums represent real capital and real footprint, which is hard to justify for one-off or short-run work.
• Visibility. The arc is completely hidden under the flux, so the operator cannot watch the pool. Troubleshooting relies on readings, bead appearance after the fact, and NDT rather than live observation.

The Defects SAW Is Most Prone To

Two defects dominate SAW troubleshooting: slag inclusions and lack of fusion. Both usually trace back to the same three culprits — incorrect parameters, contaminated flux, or poor joint preparation. Slag gets trapped when a pass is laid over an incompletely removed slag layer, when flux is not fully cleared from a narrow groove, or when bead geometry leaves a crevice the next pass cannot wash out. The fix is methodical: chip and wire-brush every pass, reshape crowns that are too tall or too concave, widen the bevel slightly if flux will not clear, and adjust travel speed so each bead wets into the previous one.

Lack of fusion occurs when the arc simply does not carry enough energy into the joint side or root: voltage too high, travel speed too fast, current too low, or a gap and bevel that do not match the procedure. Correcting it means dropping travel speed, reducing voltage slightly to narrow the arc, raising current within the procedure range, and verifying the joint dimensions against the WPS before the next run. On multi-pass joints, adding a small stringer bead along each side of the groove often restores fusion faster than simply dumping in more amps.

Porosity is the next most common complaint. It frequently comes from rust and oil on the plate, moisture in the flux, or inadequate flux coverage that lets the arc blow out momentarily. Surface contaminants are so often the root cause that good shops treat pre-weld cleaning as non-negotiable; the same discipline applies whether you run SAW or an aluminum process, as covered in this aluminum cleaner for welding prep guide. Other porosity fixes include drying recovered flux before reuse, checking hopper feed rates so coverage never thins out, and confirming gas-producing elements in the base metal stay within limits.

Cracking deserves special attention because it is the defect that reaches the customer. Excessive sulfur or other trace elements in the base metal or wire can cause hot cracking in the solidifying bead, and the fix is a combination of cleaner materials, properly matched filler chemistry, and bead sizing that avoids overly large, slow-cooling pools. Cold cracking in the heat-affected zone points toward hydrogen control: bake or store low-hydrogen flux correctly, clean moisture from bevels, and manage preheat and interpass temperature as the procedure specifies.

The practical takeaway is that SAW rewards discipline. It will produce exceptional welds day after day on thick, prepared material in the flat position, and it will punish shortcuts just as reliably. Budget for cleaning, flux management and fit-up, keep a written procedure with sensible parameter ranges, and inspect early. Do that, and the limitations above stop being surprises and become part of your standard process control.

SAW Variants and Related Processes

Ask five welders to describe submerged arc welding and you’ll get five versions of the same picture: bare wire, granular flux, hidden arc. What changes in the field is how that picture is arranged. Wire count, current type, head movement, flux delivery and consumable form all branch into variants, each built for a specific production problem. Knowing the family tree helps you recognize which setup a shop is running and why it was chosen over a relative such as flux-cored arc welding or electroslag welding.

Single-Wire DC and Single-Wire AC

The simplest configuration is one wire and one power source. Single-wire DC is the default for most fabrication because it delivers the deepest, most stable penetration for a given current, which makes it ideal for root passes and single-pass butt welds on plate. The trade-off is magnetic arc blow: as the weld pool grows and geometry becomes less symmetrical, the DC arc wanders, particularly in fillets and heavily tacked assemblies.

Single-wire AC solves that. Reversing polarity many times per second cancels the unbalanced magnetic field, so the arc stays put where DC would deflect. According to Xometry’s SAW overview, AC setups are favored for thick-plate work and multi-wire systems where arc blow would otherwise force costly rework. AC gives up some penetration compared with DC, which is why many shops run DC for the root and AC for fill and cap layers.

Twin-Wire, Tandem and Multi-Wire Heads

When one wire can’t deposit fast enough, shops add more. A twin-wire head feeds two wires into a single arc pool, usually off one power source, roughly doubling deposition without doubling travel length. A tandem setup goes further: two, sometimes three, wires run in series along the direction of travel, each on its own power source at slightly different voltage and stick-out. The leading wire punches deep penetration; the trailing wires build the bead and smooth the profile. The Hobart Brothers productivity guide frames this as the core economic argument for multi-wire SAW: higher travel speed at equal or better quality per pass, so long shipyard or pipe-mill seams close in fewer passes.

Series twin is a hybrid where both wires share one AC source in a series circuit, giving a deep, narrow profile with less arc blow than double-DC. Broadside arrangements place wires side by side across the weld direction, widening the bead for surfacing and cladding rather than deep grooves. Oscillating heads sweep a single wire across the joint to fill wide grooves in fewer layers, which pairs well with the AWS practical guide to SAW advice of matching bead width to current instead of forcing one pass to fill an oversized gap.

Narrow-Gap SAW

Narrow-gap SAW uses a near-square joint, often a 10–15° included angle, with small tightly controlled passes to weld very thick sections using a fraction of the filler metal a wide V-groove would demand. Heat input and distortion both drop. The price is precise fit-up and accurate head tracking: one misaligned pass and sidewall fusion is gone.

Strip Cladding and Special Consumables

Strip cladding swaps round wire for a flat strip electrode, typically stainless or nickel alloy, and lays a wide, shallow bead onto carbon steel plate. Dilution stays low so the cladding retains its corrosion resistance, making this the standard way to line pressure vessel inner surfaces, reactor shells and heat-exchanger components. Electroslag strip cladding is a common variant here, using resistance heating rather than an arc for even lower dilution.

Consumables branch too: metal-cored wires boost deposition and improve wetting over solid wire, while flux-cored SAW wires carry their own shielding ingredients for harder, tougher alloys. These are cousins of the alloy steel electrode families used in stick and FCAW work, selected by required mechanical properties rather than convenience. Flux delivery varies as well, from gravity hoppers to vacuum-jet systems that hold flux depth steady at high travel speeds.

Electroslag Welding: The Process SAW Gets Confused With

Electroslag welding (ESW) often shares the shop floor with SAW, and the two look similar — wire feed, flux, molten pool — but the physics differ. ESW starts with a conventional arc that melts the flux into a conductive slag; after that the arc goes out and resistance heating of current passing through the molten slag does the work. There is no sustained arc. ESW is also position-restricted in the opposite way from SAW: it runs vertical, on very thick joints roughly 25 mm and above, using a water-cooled copper shoe to contain the pool. For flat-plate seams SAW wins; for welding a 100 mm press in a single pass, ESW is in a class of its own.

How SAW Compares with Related Processes

Process	Penetration	Deposition rate	Positions	Best-fit use
SAW (DC)	Deep	Very high (up to ~20+ kg/h multi-wire)	Flat/horizontal	Long plate seams, pressure vessels
SAW (multi-wire)	Moderate–deep	Highest of the arc processes	Flat/horizontal	High-volume shipbuilding, pipe mills
FCAW	Moderate	Moderate	All positions	Field welds, structural steel
GMAW	Moderate	Moderate	All positions	Thin-to-medium plate, sheet
ESW	Very deep	Very high	Vertical only	Very thick single-pass vertical joints
Laser/hybrid	Deep, narrow	Low–moderate	Most (position-dependent)	Precision, thin sections, automated laser cells

None of these replace each other outright. SAW owns the flat, long, thick, high-volume seam. Flux-cored arc welding owns the out-of-position field joint. GMAW owns speed on thinner material. ESW owns the massive vertical joint. Shops that understand those lanes rarely buy the wrong machine — and the operators who understand them rarely get stuck with rework.

Safety, Health and Welding Codes: Protecting the Welder

The hidden arc is SAW’s greatest selling point and its most misleading one. With no visible flash and almost no spatter, it is easy to assume the operator is standing somewhere calm. He is not. A submerged arc still runs at hundreds of amperes, still throws intense ultraviolet and infrared radiation the moment flux is disturbed, and still produces a fume plume carrying the same health burden as any other fusion process. The danger is quieter, which is exactly why it gets ignored.

The Hazards That Survive the Flux

Six hazard categories account for most SAW incidents and exposures:

• Optical radiation. The arc is buried in normal operation, but UV and IR escape during arc initiation, at the start and stop of a bead, when the flux bed is too shallow, or when cover flux is knocked aside. A brief look is enough to produce photokeratitis, the familiar “arc flash.” Screens still matter around a welding tractor.
• Electric shock. SAW runs at high current on voltages that are low at the source yet capable of sustaining an arc through damp clothing, wet plate, or a grounded fixture. Wet flux and rainy outdoor workups raise the risk sharply.
• Burns. Molten slag that releases spontaneously, hot chips from slag removal, and residual heat in the workpiece cause most reported injuries. Slag peeling deserves the same caution as chipping.
• Noise. Not from the arc, but from grinders, pneumatic chippers, flux recovery vacuums, and the plant around them. Long shifts add up.
• Ergonomics. Setting tractors, kneeling at joints, and hand-holding recovery hoses for hours produces strain injuries that rarely appear in incident reports but end plenty of careers.
• Inhaled fumes. Flux decomposition releases fluorides and manganese, and when stainless or alloy steel is welded, hexavalent chromium enters the plume. Both the EU-OSHA welding fume guidance and industry reviews identify respirable particulate and metal oxides as the primary long-term respiratory risk.

What OSHA Actually Requires

OSHA’s welding, cutting and brazing fact sheet lays out the baseline in plain language: control exposure through engineering and administrative means before relying on personal protective equipment. The key regulation for the process itself is 29 CFR 1910.252, which requires adequate ventilation for welding, cutting and brazing and, where adequate ventilation cannot be maintained, requires air supply helmets or other approved respirators in confined or enclosed spaces. In an open shop that means local exhaust ventilation positioned at the weld, supplemented by supplied-air respirators when ventilation alone cannot hold the exposure down.

For stainless and other chromium-bearing alloys, the numbers get specific. OSHA’s permissible exposure limit for hexavalent chromium is 5 µg/m³ as an 8-hour time-weighted average, and it applies regardless of how little stainless passes through a SAW station. Welding procedure documents and exposure assessments should therefore name the alloy and the control method in advance. Hazard communication obligations also apply: operators and nearby workers must have access to safety data sheets for fluxes and wires, with the specific respiratory hazards of each flux stated on the sheet rather than buried in a generic category.

Codes and Procedure Qualification

Safety practice only holds up in a shop if it is written into a qualified procedure. Several code families govern this in parallel:

Code	Scope	What it controls
AWS D1.1	Structural steel	WPS, preheat, visual acceptance, welder performance qualification
ASME BPVC Section IX	Pressure equipment	Procedure and welder qualification by PQR, essential variables
AWS A5.17 / A5.23	SAW consumables	Wire and flux classification, like EM12K and EH14
CSA W59	Canadian structural welds	Design, workmanship, inspection criteria
Shipyard rules (ABS, DNV, class society)	Marine fabrication	Joint details, NDT levels, traceability of consumables

In practice, the welding procedure specification fixes the parameter window from the earlier section: current, voltage, travel speed, stick-out, flux depth, and the qualified wire/flux combination. Changing classification or jumping outside essential variables means requalification, not improvisation. Welder qualification follows the same discipline, and formal training such as an aluminium welding course is a reasonable model for how SAW operators should be certified: written theory, hands-on demonstration, and a documented performance test that expires on a schedule.

Consumable handling deserves the same rigor. Selecting a flux and wire pairing is a metallurgical decision as much as a procurement one, and the safety sheet that accompanies materials like alloy steel welding electrodes should be read before first use, especially the fluoride content and storage requirements for bonded fluxes. Reclaimed or dried flux must be tracked separately from virgin material.

Finally, inspection closes the loop. Visual acceptance under AWS D1.1 or CSA W59 governs most groove welds, while radiographic, ultrasonic, magnetic particle, or liquid penetrant testing is specified for pressure-retaining and load-critical joints. A weld that fails NDT is a safety issue before it is a rework bill, and the corrective action almost always traces back to one of the parameters, cleanliness, or moisture controls covered earlier in this article.

Applications, Industry Trends and Getting Started with SAW

Parameter tables only matter if a process earns its place on the shop floor. Submerged arc welding has done that for nearly a century, and its strongest hold today is exactly the work that keeps economies moving: steel that is thick, long, and load-bearing.

Where SAW Actually Lives: Core Industries

SAW is the default workhorse for long, heavy groove welds in plate. You will find it in:

• Shipbuilding — hull plate seams, deck welds, and bulkhead fabrication, where long continuous butt welds must hold to classification-society rules. Our AJ marine welding guide covers how marine standards shape process selection.
• Pressure vessels and boilers — circumferential and longitudinal seam welds on heavy wall sections, the daily work of firms like those in our Alamo Welding & Boiler Works overview.
• Pipelines — long seam welds on large-diameter line pipe, often welded with multi-wire heads on travelling gantries.
• Wind towers — perhaps SAW’s fastest-growing home. Tower shell longitudinal and circumferential seams, plus heavy flange welds, demand the deposition rates only SAW provides reliably.
• Bridge girders and structural steel — built-up plate girders with long web-to-flange fillet welds.
• Rail cars, offshore structures, and heavy machine manufacturing — anything where a single joint can run several meters.

Market data backs up what fabricators see daily. The global SAW equipment market was valued at roughly $3.6 billion in 2025, with a projected compound annual growth rate of about 5.4% through 2034. Shipbuilding accounted for approximately 24.6% of equipment revenue in 2025, making it the largest application segment, with oil and gas following at around 22.5%, according to Dataintelo’s market analysis.

What Is Driving the Market Forward

• Robotics and gantry automation. Column-and-arm manipulators and travelling trams are increasingly paired with robots and PLC controls, turning a one-pass setup into a repeatable production cell with recipe-driven parameter recall.
• Narrow-gap SAW. Machining joints into narrow, steep-sided grooves cuts deposited volume dramatically — sometimes by half or more on thick sections — reducing both consumable cost and distortion.
• Sensors and real-time monitoring. Voltage, current, travel speed, and stick-out are now tracked live, with some systems using vision or thermal sensing to catch deviations before they become rejectable welds. The AWS Welding Digest’s 2025 practical guide to SAW walks through how modern parameter control supports consistent quality.
• Higher-strength steels. Wind towers and heavy structures keep pushing toward quenched-and-tempered and HSLA grades, requiring careful flux selection and controlled heat input. Our alloy steel welding electrodes guide explains the underlying selection logic.
• Sustainability. The flux blanket suppresses much of the fume and light an open arc emits, and as emissions rules tighten, that lower-fume profile becomes a genuine commercial advantage.

The process itself keeps being refined as a productivity tool, as technical literature from producers such as Hobart Brothers and reference works like ScienceDirect’s SAW overview document.

Getting Started: A Practical Checklist for Shops

Adding SAW is a system decision, not a machine purchase. Work through these steps in order:

1. Assess joint volume. SAW pays off with long seams, thick sections, or high-volume repeat work. A shop welding a few short brackets a month will never justify a tractor.
2. Choose your power source and transport. Decide between DC rectifier and AC systems, then match the welding head to a bench, column-and-arm manipulator, or travelling gantry. Our welding equipment guide covers the buying considerations that apply here too.
3. Select wire and flux class. Common carbon-steel pairings follow AWS A5.17 for bare wire and flux, with AWS A5.23 covering low-alloy classifications. Match the classification to your code and service conditions.
4. Write and qualify a WPS. Under ASME Section IX or AWS D1.1, run a procedure qualification record (PQR) before production. It is the legal backbone of the weld.
5. Train operators. Even automated SAW needs people who understand flux handling, stick-out control, and restart technique.
6. Validate with NDT. Radiographic or ultrasonic testing on critical seams, with magnetic particle or penetrant testing for surface-breaking defects, per the project code.

Why the Definition Still Holds

So, back where we started: submerged arc welding is a fusion welding process in which an arc between a continuously fed bare electrode and the workpiece is buried under a blanket of granular fusible flux, which melts to shield, deoxidize, and protect the molten pool as it solidifies. That sentence has barely changed since the 1930s, and that is the point. The flux blanket still delivers what fabricators need most: deep penetration, clean high-deposition welds, low spatter, and automation-friendly consistency on industry’s heaviest joints. As long as ships, towers, pipelines, and pressure vessels are built from thick steel plate, SAW will keep quietly welding the seams nobody gets to watch.

Frequently Asked Questions

What is submerged arc welding in simple terms?

Submerged arc welding (SAW) is an automatic arc welding process where the arc is buried under a blanket of granular flux instead of being exposed to air. The flux melts to shield the weld pool and forms a slag that protects the bead as it solidifies.

Why is it called ‘submerged’ arc welding?

It is called submerged because the arc and molten weld pool sit completely beneath a mound of granular, fusible flux. You cannot see the arc while welding, and excess flux is vacuumed up and reused afterward.

What metals can be welded with SAW?

SAW is used almost exclusively on carbon and low-alloy steels, including mild steel, chrome-moly grades, and stainless cladding via strip electrodes. Non-ferrous metals are rarely welded with SAW because of oxide and wetting issues.

In which welding positions can SAW be used?

SAW is essentially limited to flat and horizontal positions because both molten metal and granular flux rely on gravity. Out-of-position welding is not practical with the standard process.

How is SAW different from MIG and stick welding?

SAW uses bare wire and granular flux with no shielding gas or electrode coating, achieves far higher deposition rates, and is mechanized. Stick and MIG welding can be done manually and out-of-position, but at much lower productivity.

What are the main disadvantages of submerged arc welding?

Its main limitations are the flat-only position restriction, large equipment footprint, sensitivity to joint fit-up and flux contamination, and the need for flux handling and recovery. Thin sheet and field work are poor fits.

What defects are most common in SAW welds?

Slag inclusions and lack of fusion dominate, usually caused by wrong voltage and travel speed, contaminated flux, or poor joint preparation. Porosity appears when bonded flux picks up moisture or plate surfaces are oily.

Is submerged arc welding safe to run without seeing the arc?

The covered arc reduces UV exposure and spatter, but hazards remain: electric shock, hot slag burns, and flux fumes containing fluorides and manganese. Ventilation, PPE and OSHA-compliant exposure controls are still required.

Which standards govern SAW work?

Filler metals are specified under AWS A5.17 (carbon steel) and A5.23 (low-alloy steel); structural work follows AWS D1.1 or CSA W59, and pressure equipment follows ASME Boiler and Pressure Vessel Code Section IX.

What industries use submerged arc welding the most?

Shipbuilding, pressure vessel and boiler fabrication, pipelines, wind towers, bridge and structural steel, rail cars and offshore structures. Shipbuilding is the largest single application segment for SAW equipment.