Submerged Arc Welding (SAW): The Complete Guide
Table of Contents
A shipyard welder once joked that his favorite part of submerged arc welding was that he could never see the arc — and that’s exactly why thousands of tons of ship hull, pressure vessel, and heavy plate get joined every year with a process most people have never watched. If you have ever wondered what ‘SAW’ stands for and why fabricators swear by it for long, thick welds, this guide answers every question in one place.
What Is Submerged Arc Welding (SAW)?
If you search “SAW” online, half the results are about saws — the cutting tools. That confusion follows the process around, so let’s clear it up immediately. In welding, SAW stands for Submerged Arc Welding, and it has nothing to do with a toothed blade. It is a fusion welding process in which an electric arc is struck between a continuously fed consumable electrode and the workpiece, completely submerged under a blanket of granular fusible flux, as the CWB Group explains. You never see the arc itself — only a mound of grey granules crawling along the joint, with a smooth, glowing bead of metal trailing behind it.
That buried arc is the whole trick. The arc melts the electrode, the base metal, and the flux together. The flux becomes a molten slag that floats over the weld pool and shields it from oxygen and nitrogen in the air, so the finished weld comes out clean, sound, and remarkably consistent.
Why the Name Confuses People
The word “submerged” is literal: the arc is submerged in flux, not in water or under a workpiece. And because the arc is buried, there is no visible arc flash. Welders working beside a SAW operation don’t need to dodge the same intense UV glare produced by open-arc processes, and the process generates very little fume compared with stick or flux-cored welding. Those two properties — no arc flash, low fume — are big reasons fabricators like running SAW in production cells where people work nearby all day.
Quick Note: SAW is almost always mechanized or automated rather than hand-welded. A welding tractor or traveler carries the head at a steady speed while the wire feeds automatically, which is exactly what makes it ideal for long, continuous welds on heavy plate.
How the Fundamental Process Works
Strip the process down to its essentials and SAW follows a simple loop:
- Granular flux is poured ahead of the weld, covering the joint completely.
- A bare consumable wire electrode feeds through a contact tip and into the pile.
- An arc is struck beneath the flux and stays buried there.
- The arc melts the wire, the plate edges, and the surrounding flux.
- Molten slag floats on top of the pool and shields it as it solidifies.
- Cool slag peels off, and unused flux is vacuumed up for reuse.
That last step matters more than it sounds. A large share of the granular flux never melts, so it gets recovered and recycled — one of the reasons SAW consumable costs stay low on long production runs. The flux does three jobs at once: it conducts the arc current into the joint, shields the molten metal, and adds alloys that fine-tune the weld’s chemistry and mechanical properties.
A Process Born in the 1930s
SAW is not a new idea. The process was developed in the mid-1930s, originally for welding the seams of storage tanks and pressure vessels where long, uniform, X-ray-quality welds were needed and manual welding was slow and inconsistent. Almost a century later, the same core principle powers ship hull panels, wind tower sections, railroad tank cars, and structural plate girders. The hardware has evolved — programmable inverter power sources, tandem multi-wire heads, automated flux recovery — but the buried arc under a blanket of flux is unchanged.
The Standards Behind SAW Consumables
What keeps all that machinery interchangeable is standardization. In the United States, the American Welding Society publishes the governing filler metal specifications: AWS A5.17 for carbon steel SAW wires and fluxes, and AWS A5.23 for low-alloy and high-alloy (including stainless) SAW consumables. Each classification on the packaging tells you the wire chemistry, the flux type, and what the combination can do in terms of tensile strength, position, and notch toughness. Matching the wire and flux as a tested pair — rather than guessing — is how you get weld metal with predictable properties.
Note: If you’re choosing consumables for the first time, read the classification code before the brand name. A wire described as an AWS A5.17 classification tells you exactly which flux family it was qualified with — useful context if you’re also comparing options like alloy steel welding electrodes for other processes.
So, to answer the principal question early and plainly: submerged arc welding is a mechanized fusion welding process that hides a continuously fed arc under granular flux, letting fabricators lay down deep-penetrating, high-deposition welds with no arc flash, minimal spatter, and very little fume. Everything else in this guide — equipment, flux chemistry, parameters, defects, and process comparisons — builds on that one idea.
How the SAW Process Works, Step by Step
From the outside, submerged arc welding looks almost too quiet to be real. No sparks, no glare, no crackle. A tractor crawls along a seam, wire spools off a reel, and a stream of grey granules pours from a hopper. Underneath that unremarkable pile of powder, however, a high-current arc is melting steel continuously, laying down bead after bead at rates stick welding never reaches. The whole process is a choreographed sequence, and once you understand the order of events, most SAW troubleshooting becomes far less mysterious.
The Sequence, From Wire Feed to Finished Bead
- Joint preparation and setup. The parts are fitted, tacked, and aligned, the plate is cleaned of mill scale and oil, and the welding head is positioned over the start of the seam. Because the operator cannot see the weld pool once travel begins, this is the stage where everything is decided: joint gap, bevel angle, wire alignment, stick-out, voltage, current, and travel speed all have to be right before the arc is struck. Guesswork here shows up later as porosity or lack of fusion.
- Flux delivery. Granular flux feeds from a hopper and drops directly ahead of and around the electrode, building a loose blanket that completely buries the wire tip. Nothing about the arc is visible at this point; the flux mound is the reason the process is called “submerged.”
- Wire feed and arc initiation. The bare filler wire is fed continuously into the joint. When it touches the workpiece, the circuit shorts and the arc restarts, still trapped under the powder. The wire itself carries the current, and typical SAW wire diameters run from 1.2 mm up to 6 mm, with welding currents reaching the high amperage ranges used for heavy plate work, according to ESAB’s process documentation.
- Conduction through molten flux. This is the part that surprises people. The granular blanket is not inert. The inner layer of flux melts into a conductive molten pool that completes the electrical path between the electrode and the workpiece, so current flows even though the wire rarely touches bare metal. That molten zone is what sustains a stable arc at currents that would blow apart an open arc process.
- Slag formation and shielding. The melted flux floats on top of the molten weld pool and solidifies into a glassy slag. It blocks oxygen and nitrogen from the atmosphere, holds heat in so the metal flows and fuses properly, and keeps spatter at essentially zero.
- Flux recovery. As the head moves on, an undercutting nozzle and vacuum hopper vacuum up the unfused granules ahead of the solidified bead. That recovered flux can usually be sieved and reused, which quietly cuts consumable cost on long runs. Only the fraction that actually melted becomes waste slag.
- Cooling, slag removal, and inspection. The bead solidifies under its own slag blanket. Once cool, the crust lifts or chips off, and the finished weld is checked for profile, penetration, and soundness.
What Flux Actually Does: Conductor, Shield, and Alloy Source
Flux is doing three jobs at once, and each one matters:
- Electrical conductor. The molten flux layer carries current from electrode to plate, which is what allows such extreme amperage without a violent, unstable arc.
- Atmospheric shield. As the mineral coating decomposes it releases a protective gas blanket and forms the liquid slag layer, keeping nitrogen and oxygen away from the hot metal. In this sense SAW is self-shielded, much like flux-based processes used in aluminum welding, but without needing an external gas cylinder.
- Alloying and deoxidation source. Flux minerals contribute silicon, manganese, and other elements to the deposited metal and scavenge oxygen out of the puddle. Change the flux and you change the chemistry of the weld, which is why matching filler and flux chemistry matters so much in code work.
Quick Note: Because you cannot watch the puddle, SAW rewards good setup and punishes sloppy setup. Fabricators set and verify parameters, flux depth, and joint alignment before starting travel, then confirm them on the first coupon. Fixing a parameter mid-run usually means stopping, cleaning, and restarting.
The practical takeaway: SAW is a closed-loop system where flux, wire, and power source work as one unit. Get the sequence right and the result is a deep, clean, spatter-free weld; get any single stage wrong and the defect reveals itself in the bead. That relationship between consumables and parameters is exactly what the next section on modern SAW equipment and setup builds on.
SAW Equipment and Setup: Power Sources, Tractors, and Heads
Set up a submerged arc welding station and you notice something odd right away: there is no gun to wrestle, no shielding gas bottle to babysit, and no arc flash to shield your eyes from. What you see instead is a machine that feeds wire into a mound of granular flux and moves itself along the joint. Everything about SAW hardware is built around that idea, which is why the equipment looks closer to a small production line than a handheld tool. As the Canadian Welding Bureau’s overview of SAW puts it, the process is designed for mechanized, repeatable welding rather than manual manipulation.
Power Sources: Why Constant Voltage Rules
Almost every SAW power source is a constant-voltage (CV) machine, usually a heavy-duty transformer-rectifier or, increasingly, an inverter. CV matters because it is self-correcting: if the welder picks up or drops arc length slightly, the current changes to pull the arc back to the set voltage, keeping the bead consistent over long runs. Most SAW power sources land in the 400 to 1,500 amp range, with twin or multi-wire systems drawing on two or more units stacked in parallel.
Polarity is the first real setup decision. A single-wire SAW system can be run three ways, and each behaves differently:
| Polarity | Penetration | Deposition | Best suited for |
|---|---|---|---|
| DC straight (electrode negative) | Deep and narrow | Lower | Root passes, single-pass welds on thick plate |
| DC reverse (electrode positive) | Shallower, wider | Higher | High-speed fill passes and surfacing |
| AC | Moderate, balanced | Moderate to high | Magnetized material, thick sections, multi-pass work |
AC has a practical advantage beyond the numbers: it eliminates arc blow, the magnetic deflection that pushes the arc off-center when welding near heavy steel structures or magnetized plate. That is why many structural and pressure-vessel shops standardize on AC even though DC delivers slightly better penetration control.
Wire Drive, Flux Hopper, and the Welding Head
The welding head is the business end of the system. It holds three things in alignment: the contact tip that transfers current to the wire, the flux delivery tube, and the vacuum recovery nozzle behind the arc. Wire is pushed from a large spool or coil through a liner and drive rolls, typically at speeds far beyond what a MIG gun manages, since SAW happily consumes several inches of wire per second.
Sitting above the head is the flux hopper, a pressurized or gravity-fed box that buries the arc under a consistent depth of granular flux. Enough flux must cover the arc at all times, usually 25 to 45 mm deep, or the arc breaks through and flashes. Behind the weld, a vacuum recovery unit sucks up unused flux for reuse while the hardened slag lifts off on its own or is chipped away.
Quick Note: Recovered flux gets screened to remove fines and slag chips before it goes back in the hopper. Reusing flux a few times is normal practice, but heavily handled or contaminated flux loses performance and can introduce porosity.
Travelers, Tractors, and Multi-Wire Heads
For flat and horizontal work, the head rides on a motorized travel carriage called a traveler or welding tractor, guided by a track, a laser seam follower, or a water-cooled copper backing bar. This is where productivity multiplies. A twin-wire head shares one arc pool, and tandem systems place two or more heads in series, each with its own power source and wire. Production configurations of twin-wire and tandem SAW can reach roughly 60 lb per hour of deposited metal, an output no manual process approaches. The Miller Welds knowledge hub on SAW deposition walks through why those numbers change with wire diameter, polarity, and flux type.
Inverter Systems and Programmable Controls
Modern SAW machines are essentially computers with heavy transformers attached. Inverter-based sources such as Lincoln Electric’s submerged arc equipment line let the operator program voltage, current, wire feed speed, and travel speed as a stored schedule, then let the machine adjust on the fly for line-voltage changes. Lincoln’s Power Wave AC/DC 1000 SD power source is a good example: it is used in robotic SAW cells, including East End Welding’s power-generation fabrication work, where every pass must match the last. The AWS Welding Digest practical guide to SAW highlights this shift toward programmable, data-logged welding as central to meeting code requirements on pressure and structural work.
The practical payoff is repeatability. When voltage, current, wire feed, and travel speed are locked into a program, a second shift welds the same joint the first shift did. That consistency matters more than any single machine brand, and it is the real reason fabricators invest in Airgas Welding Equipment-style integrated systems rather than piecing together mismatched hardware.
Before buying anything, map the work first: material thickness range, joint type, production volume, and whether the job is flat-only or needs position flexibility. Equipment that suits one shop’s long seam welds may be badly overbuilt for another’s short tack-and-fill work.
Flux and Filler Wire: Choosing the Right Consumables
Submerged arc welding runs on two consumables: a granular flux and a bare solid wire. The power source only delivers current. The flux decides shielding, slag behavior, hydrogen level, bead shape and much of the final weld chemistry, while the wire supplies the bulk of the metal. Where a stick electrode carries a coating, SAW buries the arc under a bed of flux that performs the same shield-and-slag job at far larger scale, and the same underlying idea shows up in aluminum flux welding, even though the chemistry there is a different world.
Neutral Flux vs Active Flux
A neutral flux is formulated to transfer nothing into the weld pool on purpose, so the deposit’s chemistry comes almost entirely from the filler wire. That stability from heat to heat is exactly what code work such as pressure vessels, structural plate and pipe wants from a procedure, and the CWB Group’s overview of submerged arc welding points to that repeatability as one of the process’s defining advantages.
An active flux deliberately carries alloying elements (manganese and silicon in carbon-steel grades, chromium, molybdenum or nickel in stainless and low-alloy grades) that move into the puddle as the flux melts. Active fluxes let you reach a required deposit chemistry under heavy dilution without buying an alloyed wire. The trade-off is that the transfer rate shifts with current, travel speed and flux depth, so both chemistry and bead shape wander more. Pairing an active flux with an already-alloyed wire is a classic way to overshoot specification.
Flux chemists also describe consumables by a basicity index. Acidic fluxes (below roughly 1) are inexpensive, penetrate deeply and run fast, but leave more oxygen in the deposit. Basic fluxes (roughly 1.5 and above) produce the cleanest, toughest, lowest-hydrogen weld metal. Neutral fluxes sit between the two, which is another reason they are the default when a procedure specification names a required chemistry.
Reading the AWS A5.17 and A5.23 Labels
Carbon-steel SAW consumables fall under AWS A5.17 and low-alloy consumables under AWS A5.23. Both standards classify flux and wire separately, then qualify them as a tested pair, an approach outlined in this practical guide to submerged arc welding from AWS Welding Digest. A typical label reads F7A2-EM12K:
| Element | What it tells you |
|---|---|
| F | The product is a flux intended for submerged arc welding. |
| 7 | The weld deposit must meet a minimum tensile strength of 70,000 psi (the figure is stated in units of 10,000 psi). |
| A2 | The diffusible-hydrogen category the flux must satisfy in the as-welded condition. A lower number means a drier, more hydrogen-controlled flux. |
| EM12K | The wire: E for electrode, M for the mild-steel weld-metal class, a two-digit figure tied to required strength and chemistry, and K for a potassium-bearing arc stabilizer that smooths arc action. |
For automated cells, 3/32-inch solid electrodes such as Lincolnweld LA-90, LA-92 and LA-93 are common picks, always run with a flux carrying a compatible AWS listing. For chrome-moly, manganese-moly and nickel-bearing work, AWS A5.23 does the same job, listing the wire class beside the flux qualified to test with it.
Bonded Flux vs Fused Flux
Fused flux is melted in an electric furnace and granulated into rounded, glassy grains. Those dense particles pick up little moisture, tolerate higher current densities and shrug off light rust and mill scale, which makes them the standard choice for high-speed, high-current single-pass work. Bonded flux is mixed wet with a binder such as sodium silicate, dried and crushed into irregular porous grains. It costs less and is easier to formulate in small custom batches, but those pores act like a sponge. Exposure to humid air means hydrogen pickup, and hydrogen means porosity and, in hardenable steel, cold-cracking risk.
Quick Note: Store bonded flux sealed, keep the container off the floor, and re-bake on the manufacturer’s schedule before use. Never return sweepings or contaminated flux to the hopper. When it is time to restock, a welding supplies buyer’s guide helps you compare what suppliers actually ship and how consumables are packaged.
Iron Powder: Cheap Deposition
Some fluxes come loaded with iron powder, and some shops inject it directly into the joint. Because part of the filler metal arrives as powder instead of through the wire, deposited metal per pass climbs without extra arc current. ESAB’s submerged arc welding handbook puts numbers on it: adding iron powder to the weld joint can raise SAW productivity by almost 50 percent compared with conventional SAW. The catch is that the bead runs wider and shallower, so heat input and travel speed need rechecking against fusion limits, and slag release can suffer if the flux chemistry does not match the powder.
Matching Wire and Flux the Right Way
The rule for AWS A5.23 work is simple: use wire and flux combinations that were tested and published together. Mixing brands casually may give you a sound-looking bead with unknown toughness, chemistry or hydrogen performance. The same selection discipline applies to alloy steel welding electrodes in stick processes: match the filler to the base metal and the service condition first, then worry about deposition. Good consumable handling (sealed containers, dry storage, scheduled re-baking) remains the cheapest quality control in any SAW cell.
SAW Welding Parameters That Actually Matter
When it comes to submerged arc welding, dialing in your settings can mean the difference between a clean, high-strength weld and a costly repair. The process is known for consistency, but final results are dictated by a handful of key parameters. These are the ones with the largest impact on bead quality and productivity.
1. Welding Current
The most influential setting in the process is welding current, which controls penetration depth and electrode melt rate. The goal is running the highest amperage achievable without introducing defects.
More amperage drives the arc deeper into the base metal, which suits the thick plates common in heavy fabrication. Push too far past your optimal range and you risk burn through or an irregular bead. That matters when heat input must be precise, such as when using alloy steel welding electrodes on critical joints.
2. Arc Voltage
While current controls penetration, arc voltage controls bead width and profile. Higher voltage produces a wider, flatter bead; lower voltage produces a narrower, more convex one. Excessive voltage, however, can undercut the weld toe and create a stress point for future failures.
Find the sweet spot for your joint design. In most single-wire applications, small voltage tweaks refine wetting at the toes without sacrificing fusion, and keeping voltage stable is one of the easiest ways to prevent cosmetic and structural flaws.
3. Travel Speed
Travel speed is the pace at which the welding head moves across the joint, and it directly affects heat input and bead shape. Too slow and you get excessive reinforcement, a larger heat-affected zone, and possible slag entrapment. Too fast and the arc lacks time to penetrate, leaving a lack of fusion.
Quick Note: Heat input is a balancing act. When adjusting travel speed, compensate with current and voltage to maintain proper fusion without overstressing the material.
4. Stick-Out (Electrode Extension)
Stick-out, or electrode extension, is the length of wire protruding past the contact tip before it enters the arc. It heavily influences resistive heating: a longer stick-out preheats the wire, raising melt-off rate while lowering amperage at the arc if voltage is held constant.
This underpins one of the most productive variations of the process, covered below. On standard setups, many fabricators keep stick-out between 1 and 2 inches, depending on wire diameter and power source.
5. Polarity
Polarity shapes where heat is directed in the arc. For single-wire SAW, Direct Current Electrode Positive (DC+) is the most common choice because it gives deep penetration and a stable arc. Direct Current Electrode Negative (DC-) suits select jobs where shallower penetration and a higher deposition rate are wanted, typically surfacing.
Alternating Current (AC) appears in specific multi-wire setups to help combat arc blow. Choosing the right polarity is vital and is usually recommended by your filler metal and flux combination.
6. Flux Depth
Unlike open-arc processes, SAW relies on a blanket of granular flux to shield the weld, so the layer must fully submerge the arc. Too little flux exposes the arc, causing porosity and spatter; too much restricts visibility of the puddle, increases drag, and can destabilize the arc. A general guideline is a layer roughly 0.75 to 1.5 inches thick, depending on amperage. Consistent coverage matters as much as your electrical parameters, especially on automated runs.
The Long-Stick-Out Technique
Arguably the biggest productivity booster for a single-power-source setup is the long-stick-out process. Extending the electrode well beyond standard length lets resistive heating preheat the wire before it reaches the arc, which raises the melt-off rate dramatically without pushing welding current to extreme levels.
That translates directly into deposition rates. A widely accepted rule of thumb from ESAB states that for single-wire DC+, the deposition rate in kg/h rises with current and falls with larger wire diameter: Dep. rate = Amp / (50 × Diam^0.3). Used together with the long-stick-out technique, this lets fabricators boost output while keeping heat input manageable.
Short Note: When implementing long stick-out, make sure your wire drive and contact tip can accommodate the extended length to prevent feeding issues. Precision here is key!
| Parameter | Primary Effect | Excessive | Insufficient |
|---|---|---|---|
| Current (Amps) | Penetration & melt rate | Burn through, distortion | Lack of fusion, low deposition |
| Voltage (V) | Bead width & profile | Undercut, wide flat bead | Convex bead, poor wetting |
| Travel Speed | Heat input & shape | Lack of fusion, ropey bead | Overlap, slag inclusions |
| Stick-Out | Preheat & deposition | Arc instability, poor control | Lowered deposition, less preheat |
| Polarity | Heat distribution | Varies by setup | Improper penetration |
| Flux Depth | Shielding of the arc | Instability, drag | Porosity, spatter, flash |
Deposition Rates and Productivity: Why Fabricators Love SAW
Ask a shop foreman why they switched to submerged arc welding and metallurgy rarely comes up first. You hear a number: pounds per hour. SAW’s reputation rests on one simple advantage — it dumps metal into a joint faster than almost any other arc process, while a machine does the traveling so the operator focuses on control rather than endurance.
What Deposition Rate Actually Means
Deposition rate is the weight of filler metal placed into the weld joint per unit of time, usually in pounds per hour. It is not travel speed, and it is not deposition efficiency — the share of melted wire that actually stays in the weld rather than flying off as spatter. Keeping those three numbers straight is the foundation of any honest productivity comparison, as The Fabricator’s breakdown of deposition rate, efficiency, and output explains.
A single-wire SAW system can achieve deposition rates of up to 40 lb/h. A highly skilled manual SMAW welder, by contrast, typically deposits around 3 to 5 lb/h — and that welder also spends time changing electrodes, chipping slag, and grinding. Miller’s technical overview of SAW deposition rates frames this gap as the single biggest reason heavy fabrication shops automate.
Deposition Efficiency: The Quiet Multiplier
Raw deposition rate is only half the story. Deposition efficiency — sometimes called transfer efficiency — is the percentage of consumed filler metal that actually becomes part of the weld. SAW approaches 99 percent because the granular flux blankets the arc and virtually eliminates spatter. SMAW sits near 55 percent, since a meaningful share of the consumable is lost to fume, spatter, and the stub you throw in the scrap bucket.
Quick Note: A process depositing 20 lb/h at 99 percent efficiency delivers more real weld metal than one depositing 24 lb/h at 70 percent. Always compare effective deposition, not just the headline number.
That efficiency compounds with iron-powder-bearing fluxes, which can push deposited metal above the wire’s own melt-off rate. It is one reason consumable selection — covered in Alloy Steel Welding Electrodes: Selection, Use & Safety — matters as much as the power source.
Operating Factor: Where Manual Welding Really Loses Time
Even excellent deposition rates mean little if the arc is off. Operating factor is the percentage of a shift the arc is actually burning. The industry average for manual welding is roughly 20 percent — about 12 minutes of arc-on time per hour, according to The Fabricator. The other 48 minutes go to fit-up, electrode changes, interpass cleaning, repositioning, and breaks.
SAW flips this equation. A welding tractor travels at a set speed, wire feeds continuously from a large spool, and flux feeds from a hopper, so operating factors climb well above manual levels. Fewer stops also mean fewer arc starts, which is where many defects begin.
Scaling Up: Twin-Wire and Multi-Wire Systems
Single-wire SAW is impressive; multi-wire configurations are where productivity becomes extraordinary. Twin-wire setups can reach approximately 60 lb/h, and production-oriented tandem, series, or three-wire systems go higher still. EWI’s productivity guidance on SAW and Hobart Brothers’ SAW productivity article both describe multi-wire heads as the standard route for heavy plate, longitudinal seams, and large-diameter pipe.
Adding wires usually means adding heat, so joint design and distortion control must be planned. Still, on a 25 mm butt groove at a plate mill, the gap between one pass and a properly sequenced multi-wire program is measured in hours, not minutes.
How SAW Compares on Paper
| Process | Typical deposition rate | Approx. deposition efficiency | Position flexibility |
|---|---|---|---|
| SMAW (stick) | 3–5 lb/h | ~55% | All positions |
| GMAW / MIG | 5–12 lb/h | ~90–95% | Most positions |
| FCAW | 5–15 lb/h | ~80–90% | Most positions |
| SAW (single wire) | Up to 40 lb/h | ~99% | Flat and horizontal only |
| SAW (twin wire) | ~60 lb/h | ~99% | Flat and horizontal only |
Notice the trade: SAW wins on speed and efficiency but surrenders position flexibility, which is why you will not see a SAW rig overhead on a jobsite. For comparisons with wire processes in aluminum work, see our guides to Aluminum Flux Core Welding: What Actually Works in 2026 and Aluminum MIG Welding: The Complete Practical Guide.
Short Note: Productivity is not just pounds per hour. Repositioning the workpiece often costs more than the welding itself, so shops routinely rotate entire assemblies so every weld can run flat. As the CWB Group notes, SAW’s flat-position restriction is real but rarely fatal — smart shops design the sequence around it.
The Hidden Savings
The economics go beyond metal in the joint. Lincoln Electric’s SAW equipment overview highlights consistent, repeatable weld profiles that cut grinding and inspection rework, plus flux that can be recovered, screened, and reused to trim consumable cost per pound. Add reduced fume compared with open-arc processes, lower operator fatigue, and easy automation, and the per-joint cost of SAW on long, heavy welds is difficult to beat.
The bottom line: SAW is not the fastest way to weld a bracket or a fence post. But when the job is long, thick, and flat, no common arc process turns kilowatts and wire into finished weld metal faster.
Where SAW Is Used: Applications and Joint Types
Nobody picks submerged arc welding because it is exciting to watch. Fabricators pick it when the drawing calls for a three-metre seam in 25 mm plate and a stick-welded equivalent would eat a crew’s entire shift. As the AWS Welding Digest practical guide to submerged arc welding puts it, SAW shines where long welds and thick materials are the norm. That single sentence explains almost its entire industrial footprint.
Industries and Assemblies Built With SAW
Pressure vessels and boilers. Shell seams, head-to-shell welds and nozzle attachments are classic SAW territory. Plate is rolled on turning rolls, the head sits on a boom or column-and-boom manipulator, and the weld is made in the flat position where SAW is most comfortable. Fabricators such as Alamo Welding & Boiler Works build their boiler expertise around exactly this kind of heavy, radiographically inspected seam work.
Shipbuilding. Hull panel lines, stiffener-to-plate welds and deck seams are long, repetitive and mostly flat. Shipyards widely use iron-powder-enhanced SAW to cut fabrication time, because the iron powder carried in or on the flux melts into the weld pool and boosts deposition without extra arc energy. Where hulls switch to aluminium superstructures or ferry decks, SAW steps aside and processes like those covered in aluminum boat welding take over, since aluminium cannot tolerate a submerged-arc-scale heat input.
Pipe seam welding. Longitudinal submerged arc welding (LSAW) and spiral pipe lines weld continuous seams at steady, high travel speeds. This application drives its own consumable niche: fluxes with particle size optimized for high-speed continuous seam welding, so the arc stays properly covered at speed and the bead runs smooth, as detailed in ESAB’s submerged arc welding reference.
Structural plate girders and bridges. Built-up I-sections for bridges and buildings need long fillet welds joining webs to flanges. Gantry-type SAW rigs travel the length of the girder in one continuous pass rather than stopping every electrode length.
Wind towers. Rolled steel cones and cylinders are welded circumferentially with internal and external SAW heads working from both sides of the seam, followed by flange welds on the machined ring forgings.
Storage tanks. Tank floors, shell courses and nozzle welds use portable welding heads that orbit the joint, giving consistent penetration on plate from 6 mm to well over 30 mm.
Locomotives and heavy equipment. Underframes, boom sections, bucket edges and drivetrain castings combine long groove welds with wear-resistant surfacing. Much of this work involves heat-treated or alloy plate, where filler selection follows standards like AWS A5.23 and the same discipline described in alloy steel welding electrodes guidance.
Power generation. Boiler drums, turbine casings and pressure pipework now frequently use robotic SAW, adopted specifically where repeatable, high-quality welds are required and operator variation must be removed from the equation. Modern Lincoln Electric submerged arc systems pair programmable travel carriages with closed-loop controls for exactly this work.
| Application | Typical joint | Why SAW fits |
|---|---|---|
| Pressure vessels | Butt groove, circumferential | Deep, clean penetration; low hydrogen; flat position on rolls |
| Shipbuilding panels | Fillet, lap, butt | Iron-powder flux boosts deposition on long runs |
| Pipe seam (LSAW/spiral) | Butt groove | High-speed continuous welding with tuned flux |
| Plate girders | T-joint fillet | Single-pass fillets up to substantial leg sizes |
| Wind towers | Circumferential butt | Two-sided heads, consistent multi-pass fills |
| Tank fabrication | Butt and edge | Portable orbiting heads, uniform bead on plate |
| Heavy equipment | Groove plus surfacing | High deposition for wear pads and hardfacing |
Typical SAW Joint Types
- Butt joints. The bread-and-butter joint: V, U, J and K bevels in plate from roughly 10 mm to several hundred millimetres, welded in multiple passes with stringer beads or oscillated weave beads.
- Full-penetration groove joints. Narrow-gap and single-V preparations cut weld volume dramatically on very thick sections, saving both wire and time.
- Fillet and T-joints. Stiffeners, cruciform connections and web-to-flange welds made in the flat position, often with iron-powder flux to reach large leg sizes in fewer passes.
- Surfacing and cladding. Wear plates, bucket liners and corrosion-resistant layers built up by overlay welding, where SAW’s high deposition efficiency pays for itself quickly.
- Edge and lap joints. Less common, but used for box sections, seam-closed cylinders and thin stiffener work.
Quick Note: SAW is essentially a flat-position process (or horizontal on circumferential seams turned on rolls). If a joint cannot be repositioned to roughly 1G, SAW usually is not the answer. Engineers routinely rotate the part rather than fight the process.
The pattern is consistent: if the weld is long, the plate is thick, the joint can be turned flat and the volume justifies setting up a tractor and flux hopper, SAW is usually the economical answer. For short, out-of-position or thin-gauge work, lighter processes remain the better fit.
Safety, Fumes, and Health Considerations in SAW
Submerged arc welding looks like the polite cousin of the welding family: no exposed arc, almost no spatter, barely visible smoke. Fabricators read that calm surface as proof of safety, and that is exactly the trap. SAW genuinely is easier on the crew than most processes, but the hazards it does carry are chronic rather than dramatic. They build up over years of quiet overexposure instead of announcing themselves with a flash or a cough on the first bead of the shift.
What Is Actually in the Fume
The arc never stops vaporizing metal, it just does it under a blanket. The plume stays low and thin until someone rakes slag or vacuums unconsumed flux while it is still hot, which is when most of the release happens. OSHA identifies welding fume health effects that include metal fume fever, kidney damage, and effects on the nervous system. Manganese draws the most concern: prolonged inhalation can produce tremor, slowness, and coordination problems that closely resemble Parkinson’s disease, and the damage does not reliably reverse after exposure stops.
Exposure Limits and Respirator Rules
OSHA’s permissible exposure limit for total welding fume is 5 mg/m³ for airborne fume produced when welding mild steel, iron, or aluminum. Treat that as the starting line for your assessment rather than the whole assessment, because it does not break out manganese, chromium, or nickel individually. NIOSH lists welding fumes in its Pocket Guide with substance-specific limits and respirator recommendations, which is where most written respiratory protection programs get their filter and cartridge choices.
When engineering controls cannot hold exposure under the limit, a formal respirator program is required: fit testing, medical clearance, change-out schedules, and training. The hardware is easy to buy through suppliers such as Airgas welding supplies; the written program has to come from your own exposure data.
Fluorides: The Warning on the Flux Bag
Every flux package carries a version of the same caution: heated flux releases fumes containing fluorides that may irritate the eyes, nose, and throat, as university EHS guidance on welding, cutting, and brazing puts it. Irritation is the short-term complaint. Sustained fluoride exposure affects the same respiratory tissue and bones that stick welders face on basic electrodes, which is why consumable guidance such as our notes on alloy steel welding electrodes treats flux handling and ventilation as part of the procedure, not an afterthought. Basic, high-basicity fluxes used on critical low-alloy work generally produce more fluoride than acidic neutral grades.
Quick Note: Never sweep dried flux or slag dust with compressed air, and never store unconsumed flux in a sealed bag while warm. Both trap moisture and turn a routine cleanup job into a dust and fume exposure. Use a HEPA vacuum and let flux cool in an open, dry container.
Beyond Fume: Burns, Heat, and Confined Spaces
The flux mound feels cool and hides molten slag underneath, so welders get burned reaching across a “finished” bead that is still glowing. Preheat on thick plate adds another burn source, along with the glowing workpiece itself. Confined spaces are the serious one: vessel interiors, tank sections, and box girders trap heat and fume, and a welder folded into a shell can exceed exposure limits in minutes even when the open-shop reading is perfectly fine. Add the ordinary shop risks, tractors and wire feeds running unattended, slag dropping through a nearby curtain, hot fits and lifts, and SAW’s quiet reputation starts working against you.
Controls That Actually Work
- Local exhaust ventilation positioned at the flux blanket and at the slag-removal station, where the real plume appears.
- Enclosure and dilution ventilation around the cell so fume never drifts toward office or break-room air.
- Air monitoring for total fume and manganese, repeated when flux, wire, or the job changes.
- Permitted confined-space entry with gas monitoring, an attendant, and retrieval gear before anyone crawls inside a vessel.
- PPE: heat-resistant gloves, leathers, safety glasses, and a face shield during slagging, plus respiratory protection per your written program.
- Housekeeping with a HEPA vacuum, dry storage for reclaimed flux, and no compressed-air blowdown.
Compared with gouging or flux-cored work, SAW hands you a real head start: the arc stays fully buried, so there is no arc flash filling the shop, and visible fume at the arc is comparatively low, as the CWB’s overview of submerged arc welding describes. Use that advantage deliberately instead of assuming the process protects people by itself. For a useful contrast, air arc welding and carbon arc gouging shows what a far fume-heavier hot-work method demands from the same crew and the same ventilation budget.
Common SAW Defects, Problems, and How to Fix Them
The biggest hurdle with submerged arc welding is its greatest strength: the arc is buried under a blanket of flux. While this protects the weld, it also hides flaws. A SAW bead can look perfectly smooth on the surface while concealing porosity or slag inclusions underneath. For this reason, troubleshooting should never rely on appearance alone. Cut, etch, or x-ray the weld to verify its integrity before signing it off for code work.
Porosity: The Flux Is Usually the Culprit
Gas pores are far more likely to stem from the consumables than the power source. Moisture absorption, rust, mill scale, oil, paint, or dirty handling can all contaminate flux and introduce gas into the weld. An excessively deep and fluid weld pool can also trap gas, giving it no time to escape before solidification.
To correct this, focus on cleanliness and small parameter adjustments:
- Store properly: Keep flux in a sealed, dry, or heated container, and re-bake it to the manufacturer’s specifications if it has been exposed to moisture.
- Clean the joint: Remove oil, rust, and any moisture from the base metal before welding.
- Adjust heat input: Reduce heat input or slightly increase travel speed to shallow the pool and allow gas to escape.
- Inspect your wire: Rusted or oily wire will feed the same issue straight into the arc.
Slag Inclusions: Clean Between Passes
For multi-pass joints, leftover slag is the leading cause of inclusions. The next pass will not melt it away, so any trapped film, ridge, or pocket becomes a permanent discontinuity, often only revealed by radiography.
The fix comes back to discipline: wire-brush or grind every pass to bright metal, tune your flux recovery vacuum to remove old slag and unfused flux, and ensure each bead overlaps the previous one. When a weld does fail inspection, carbon arc gouging is the standard route for removing the rejected material before rewelding.
Quick Note: Poor slag release often goes hand in hand with travel speed. A flux that lifts cleanly at a slow speed may hold tight and tear into shards at a much higher one. Whenever you change travel speed, run a bead-on-plate trial and check how the slag lifts before moving to production.
Undercut and Lack of Fusion
Undercut is typically caused by too much voltage for the travel speed and bead width, carving the edges faster than filler metal can fill them. Lowering voltage, raising travel speed, adding a wire, increasing flux depth, or switching to AC can all help stabilize the arc and prevent it.
On the flip side, lack of fusion at the toe or root usually means there isn’t enough energy. Lowering travel speed, increasing current, widening the bead with voltage adjustments, or preheating heavy sections will help the arc reach the joint properly.
Centerline Cracking in Deep, Narrow Beads
Longitudinal centerline cracks are a solidification cracking issue. High heat input paired with a deep, narrow bead creates a long columnar grain structure that pulls apart as the center of the pool freezes. This is especially common with low-hydrogen, lower yield strength steel plate and certain alloy compositions.
Two of the most reliable fixes are to raise voltage to widen the bead and reduce its depth-to-width ratio, or to abandon a single wide pass in favor of the stringer bead technique, placing two or more narrower, shallower beads side by side.
Rough Beads and General Bead Shape Problems
Erratic profiles, excessive reinforcement, or rough surfaces often point to unstable wire feed, arc blow, incorrect stick-out, or a mismatch between travel speed and current. The long stick-out technique can boost deposition, but when pushed too far, it will produce a rougher, narrower bead. Before blaming the flux, confirm your wire feed speed and voltage pairing against the manufacturer’s chart.
Quick Defect Reference Table
| Defect | Likely Causes | Practical Corrections |
|---|---|---|
| Porosity | Wet or contaminated flux, dirty plate, deep weld pool, rusty wire | Re-bake flux, clean joint, reduce heat input, check wire storage |
| Slag inclusion | Slag left between passes, weak vacuum recovery, poor bead overlap | Brush every pass, tune flux recovery, verify bead placement |
| Undercut | Excessive voltage, fast travel, excessive penetration | Lower voltage, slow down, add wires, deepen flux cover, or try AC |
| Lack of fusion | Too cold or too fast, narrow joint not reached | Increase current, slow travel, preheat, widen bead |
| Centerline crack | High heat input, deep narrow bead, columnar grain structure | Raise voltage to widen, use stringer bead technique |
| Poor slag release | Travel speed too high for the flux type | Test flux at production speed, adjust flux or travel speed |
Most importantly, these parameters interact. Raising current to fix fusion may deepen the pool and reintroduce cracking. Always validate your corrections with a test plate, and for code work, use the same inspection method that will be used on your production welds.
SAW vs Other Welding Processes: An Honest Comparison
Every welding process wins somewhere and loses somewhere else. Submerged arc welding is not a universal answer, so here is how SAW compares with SMAW, GMAW, FCAW, and electroslag welding, and when it is the right call.
SAW vs SMAW (Stick Welding)
The contrast could hardly be sharper. Stick welding is portable, works in any position, needs almost no fixturing, and runs from a single machine. SAW needs a clean joint, a flux supply, a wire feed system, a travel carriage or mechanized head, and a position where granular flux will sit on the weld zone. The Canadian Welding Bureau’s overview of SAW is blunt about that limit: the process is largely confined to flat and horizontal welding, because gravity decides where the flux blanket goes.
What SAW gives back is throughput. SMAW typically deposits around 3 to 8 lb per hour, while SAW routinely runs 15 to 40+ lb per hour depending on wire count, polarity, and flux. On a long butt groove in a pressure vessel shell course, that gap turns into fewer labor hours per pound of deposited metal, where real cost lives. For the stick side, see our Advantages of Stick Welding: Why SMAW Still Wins.
SAW vs GMAW and FCAW
Gas metal arc welding and flux-cored arc welding are the flexible middle ground, handling out-of-position work, thinner sections, smaller assemblies, and field repairs SAW cannot reach with far less fixturing. That lowers the entry cost for a shop running varied short jobs.
For thick-plate butt welds in production, though, SAW usually wins on both deposition and efficiency. The arc runs under flux, with no shielding gas to blow away and little spatter, so more energy and consumed wire end up in the joint. Miller’s technical note on SAW deposition rates and EWI’s productivity guidance on submerged arc welding make the same point: mechanized SAW stacks metal faster than semi-automatic processes on heavy sections, with a calmer, more repeatable bead.
The catch is that GMAW and FCAW remain the better answer for vertical, overhead, and positional work. When a job runs mostly out-of-position, start with an All Position Welding Rod Guide rather than forcing a flux blanket into a vertical groove.
Quick Note: Deposition rate and deposition efficiency are different things. Rate is pounds per hour consumed; efficiency is the percentage that ends up in the weld. SAW fluxes with iron powder additions push efficiency well above the arc-welding norm, which is why SAW output looks so strong per hour worked.
SAW vs Electroslag Welding
Electroslag welding, the other heavy-plate specialist, solves a different problem: it fuses very thick sections in a single vertical pass using a consumable guide and a molten slag bath. The trade-off is a coarse, columnar grain structure and a large heat-affected zone, so ESW joints often need temper-bead control, post-weld heat treatment, or strictly limited service where fatigue and toughness matter. SAW builds the joint in controlled passes instead, giving finer grain, better toughness, and more flexibility in heat input, which makes it the safer default for multi-pass plate work. The Fabricator’s breakdown of deposition rate, efficiency, and output shows why raw speed alone rarely decides the winner.
Cost, Setup, and Flexibility: The Real Decision Factors
- Setup investment: SAW needs fixtures, alignment, flux handling and recovery, and a travel system. Modern Lincoln SAW equipment adds programmable, inverter-based controllers that soften the learning curve, but the capital cost is real.
- Operating factor: automated SAW keeps the arc running a much larger share of the shift than hand welding, so the effective output gap is wider than the deposition numbers suggest.
- Flexibility: SMAW and FCAW travel to the work; SAW requires the work to come to the machine, or a well-built rig to go to it.
- Consumables: SAW flux storage, drying, and recovery add handling that stick, MIG, and flux-core shops do not carry.
When SAW Is the Right Choice, and When It Is Not
Choose SAW when you have long, thick butt or groove welds in flat or horizontal position, repeatable fit-up, and enough volume to justify fixturing: pressure vessels, storage tanks, ship hull plate, seam-welded pipe, plate girders, wind tower shells, and heavy surfacing work.
Choose something else when the work is out-of-position, the assemblies are small and varied, the schedule is too short to set up, or the part must be repaired in the field. A one-off bracket on a vertical column is a stick or MIG job.
The honest summary: SAW trades flexibility for automation and volume. If your workload is thick, long, repetitive, and positionally friendly, nothing matches its cost per pound of weld metal. If it is not, a simpler process is usually cheaper once you count setup time.
Frequently Asked Questions
What does SAW stand for in welding?
In welding, SAW stands for Submerged Arc Welding — a process where the arc is buried beneath a blanket of granular flux. The acronym sometimes causes confusion with the power tool, but in fabrication contexts it always means submerged arc welding.
Can you see the arc in submerged arc welding?
No. The arc is completely covered by granular flux, so no visible arc or flash is produced during normal operation. This is one of SAW’s biggest safety advantages, since it greatly reduces the risk of arc eye (photokeratitis).
What are the main advantages of submerged arc welding?
SAW offers very high deposition rates (up to about 40 lb/h for single wire), deposition efficiency near 99 percent, deep penetration, smooth bead appearance with minimal spatter, low fume generation, and excellent repeatability when mechanized.
What position can SAW weld in?
SAW is primarily limited to flat and horizontal welding positions because the granular flux must blanket the arc and gravity would otherwise let molten metal run out. Special fixtures and turning gear can rotate large workpieces to keep the weld seam in a downhand position.
What kinds of joints and materials are best suited to SAW?
SAW works best on thick steel plate in butt and groove joints, and is widely used for pressure vessels, ship hulls, storage tanks, wind towers, pipe longitudinal seams, and heavy structural girders. It can also join stainless and low-alloy steels with the right wire/flux combination.
What is SAW flux and why does it matter?
Flux is a granular, fusible material that shields the molten weld from air, stabilizes the arc, can act as a deoxidizer, and contributes alloying elements. Fluxes are classified as neutral or active, and choosing the correct flux for the wire and joint is critical to weld chemistry and bead shape.
How much does SAW deposition rate compare to stick welding?
A skilled manual SMAW welder typically deposits 3 to 5 lb of weld metal per hour, while single-wire SAW can deposit up to around 40 lb per hour. This is why SAW dominates high-volume plate fabrication.
Is submerged arc welding safe to do without fume extraction?
SAW produces relatively little visible fume because the arc is submerged, but flux heating still releases fumes — including fluoride-containing fumes that irritate the eyes, nose, and throat. OSHA ventilation, PPE, and where needed respiratory protection requirements still apply.
What are the most common SAW defects?
The most frequent problems are slag inclusion, porosity from damp flux, undercut from excessive voltage, lack of fusion, and centerline cracking from overly deep beads. Most are resolved by correcting parameters, flux storage, or travel speed.
Nicholas Harper
Editorial TeamCovering welding processes, equipment selection, and shop-floor practice for WeldZen — with every claim checked against the governing code.
Related Studies
Get The Next Benchmark First
New field studies and calibration sheets, once a week. No spam.