History of MIG Welding: From 1948 to Today’s Machines

A shop foreman asks why the machine his father called “MIG” is sold today under the name GMAW — and whether MIG welding was really invented in 1948, or a century earlier. The honest answer: MIG was less a single invention than a 150-year argument about how to keep air out of a molten weld puddle, finally solved in 1948 with a constant-voltage power source and a continuously fed wire. Once you see that story, every knob on your machine stops being mystery numbers.

Before MIG: Welding’s Long Road From Forge to Electric Arc

Long before a wire was ever fed into an arc, metal was joined by fire. For centuries, forge welding allowed blacksmiths to heat and hammer pieces together. While effective for its time, it was slow, labor intensive, and simply could not be used to join large, enclosed structures without tearing them apart to reach the joint.

The Arc is Discovered

That changed with the electric arc. In 1800, Humphry Davy demonstrated a short-pulse arc, and in 1803, Russian scientist Vasily Petrov produced the first continuous electric arc in News of Galvanic-Voltic Experiments. Davy would rediscover it in 1808, unaware of Petrov’s work. With this, a concentrated, controllable heat source hot enough to melt steel was finally a reality. However, one problem remained: molten metal left exposed to air would oxidize, creating brittle and unreliable joints.

From Carbon to Coated Electrodes

The first practical attempt came from the battery trade. In 1881, Auguste de Meritens used an arc to join lead plates. This led to his pupil, Nikolai Benardos, working alongside Stanislaw Olszewski, to patent the carbon electrode arc welding method with a British patent in 1885 and a U.S. patent in 1887. Carbon arc welding is still used today for gouging and cutting, including in processes such as air arc welding and carbon arc gouging. Still, without any form of shielding, oxidation severely limited its weld quality.

Metal electrodes proved to be the fix. Nikolai Slavyanov patented one in 1888, followed by C. L. Coffin in 1890. Around 1900, A. P. Strohmenger introduced a coated metal electrode in Britain. That coating provided far more arc stability and is the direct ancestor of shielded metal arc welding, better known as stick welding.

The Blueprint Without the Gas

Two developments of the 1920s would set the stage for MIG. First, automatic welding with continuously fed electrode wire meant welders no longer had to manually swap rods. Most importantly, in 1920, P. O. Nobel of General Electric built a forerunner to GMAW using direct current, a bare electrode wire, and arc voltage to self-regulate the wire feed. The self-regulating machine was already there. What wasn’t, were welding atmospheres to protect the puddle, so it still had to run in open air.

Progress didn’t stop there. Holslag introduced AC welding in 1919, Kyle Taylor released stud welding in 1930, and submerged arc welding followed with the arc running under a blanket of granular flux. The rise of reactive metals like aluminium and magnesium also sparked growing interest in properly preparing aluminium for welding.

The Proof That Changed Everything

All of this work culminated with the M/S Carolinian, launched in 1930 as the first all-welded merchant vessel. Built by a team of nine workmen, it saved roughly 75% of the cost and about 77,000 lb (35,000 kg) of steel rivets compared to a riveted ship of the same size. Welding won not for its looks, but for its weight savings, lower cost, and elimination of leak-prone rivet holes. By World War II, that advantage was put to use on a massive scale across shipyards, tanks, aircraft, and artillery, building the industry that remains today.

1948: The Constant-Voltage Breakthrough That Created Modern MIG

If you remember one thing about 1948, make it this: the machine held a constant voltage across the arc, not a constant current. That single choice is why a welding machine can push wire at you continuously and still lay down a stable bead. The feeders, the guns, the pulse programs and the synergic modes all grew out of it.

Getting there took a century and a half of arc welding. The long sweep of welding history handed MIG a working electric arc and an unsolved problem. The arc was solved. Keeping air away from a molten pool of metal was not.

Why a Continuously Fed Wire Needs a Self-Regulating Arc

Stick welding asks the welder to do three things at once: hold the arc, add filler, and manage heat. Put a wire feeder in charge of the adding and the machine inherits that job. The machine has to regulate.

Here is the failure mode. The feeder pushes wire toward the arc at a fixed speed, and the arc has to melt that wire at exactly the same rate. If the wire melts faster than the feeder feeds it, it melts back into the contact tip, the tip swells, the gun jams, and you stop to cut off the tip and clean the gun. That is burn-back, and in the late 1940s it was the biggest reason nobody could weld with a fed wire and expect the machine to behave.

The mirror-image failure is just as bad. If the wire arrives faster than the arc can melt it, the tip pushes into the puddle, the pool cools, the arc goes short and stiff, and the bead freezes. The joint ends with a crater where a weld should be. Both failures are one problem seen from two ends: melt rate and feed rate drifting apart. A process that cannot correct that drift cannot be handed to a machine, because a machine cannot hold a torch a millimetre closer.

What Constant Voltage Actually Does

Arc length dictates the voltage an arc needs. A longer column of plasma between the tip and the pool demands a higher voltage to sustain itself. A constant-voltage (also called constant potential) source treats voltage as fixed and lets current go wherever the arc length demands it.

The result is a self-correcting swing in heat input whenever the arc length shifts. If the arc collapses, the machine raises the current, more current melts the wire faster, and the tip burns back at the correct rate so the arc length recovers on its own. If the arc stretches, current drops, the wire melts more slowly, the wire feeds forward and the arc shortens again. Nobody touched the machine. The arc corrected itself in a fraction of a second, which is why this became known as a self-regulating arc.

Constant current, by contrast, is what your stick machine does. The source holds the amperage and the welder holds the arc. Lean in and the voltage climbs; pull back and it drops. The welder’s hands are the feedback loop. It works because a human is a superb arc regulator, but it does not scale to a robot or to an unattended feeder. Shielded metal arc welding still runs on that logic, and our breakdown of where stick welding still wins is really a list of places where manual arc control still beats machine control.

The 1948 Machine and the Name That Stuck

The breakthrough that made GMAW real in 1948 is generally credited to Battelle Memorial Institute, where the work paired a smaller-diameter consumable electrode and a continuous wire feeder with a constant-voltage power source developed by H. E. Kennedy. The GMAW history recorded in the standard reference states the trade-off plainly: the process welded non-ferrous materials fast, but only because it required expensive shielding gases. The same chronology lists gas tungsten arc welding as perfected in 1941, with gas metal arc welding following in 1948.

On the commercial side, industry accounts commonly credit the team at Hobart Brothers Company of Troy, Ohio, and engineer Dave McKee with turning the idea into a product and developing the continuously fed welding gun. Treat that as widely reported trade attribution rather than settled history, because the late 1940s were never documented with the precision the story deserves.

The economics explain why the first machines were stuck on aluminium and stainless. Argon and helium were ruinously expensive per hour, so even an impressive deposition rate produced an expensive weld. Non-ferrous work was worth that price; ordinary carbon steel was not, and it had to wait for cheap shielding gas, which is what the next chapter delivers. Our complete guide to aluminium MIG welding shows how those early applications look in practice today.

One piece of housekeeping. The trade name MIG was picked up commercially in the early 1950s, with trade histories generally crediting Airco with registering it as a brand. It stuck so hard that it outlasted its own accuracy, which is why your manual now calls the process GMAW. The full naming argument belongs to the standards section.

By the end of 1948, the welder’s job had shrunk to something recognisable: set the machine, hold the gun, and manage travel speed and gun angle. A small change in words, and an enormous change in what the process could become.

CO2, Shipyards, and the Navy: Making MIG Pay Off on Steel

MIG had a problem in 1949 that no clever power source could fix: the gas bill. The constant-voltage machine and the continuous wire feeder were ready, but the atmosphere they required was a luxury. Argon was available, and helium was used in aerospace, but cylinder charges of either pushed the per-hour cost of a simple steel fillet weld past what many fabrication shops would accept.

That is why the process stayed where the gas was already paid for. Aircraft, missile work, aluminium extrusion and stainless food equipment drew on the same supply infrastructure, and shops that specialised in those materials could justify the argon. Ordinary steel fabrication could not. If you want to see what MIG was actually doing on aluminium during that era, our aluminum MIG welding guide covers the process in depth, but the historical point stands: the gas was the barrier, not the arc.

1953: Carbon Dioxide Changes the Math

The breakthrough came from an unglamorous source. In 1953 the use of carbon dioxide as a welding atmosphere was developed, and it quickly gained popularity in gas metal arc welding because it made welding steel far more economical than an inert gas. CO2 was cheap, abundant, and produced in enormous quantities as a by-product of other industries, so a fabricator could buy shielding gas by the pallet instead of the cylinder.

Welding with it changed the character of the arc too. Pure carbon dioxide gives deep penetration and a wide, flat bead that fills a groove efficiently, which is exactly what heavy structural work wants. The trade-off is that CO2 is chemically reactive. It oxidises the molten pool, which means more spatter flying off the torch, a rougher bead, and a rougher cleanup job afterwards. A specialist stockist’s guide to welding gas and cylinders is worth a look if you are pricing shielding for a shop.

Industry settled on a compromise. Blends such as 75% argon / 25% CO2 and 90% argon / 10% CO2 became the standard working mixtures: the argon keeps the arc stable, improves wetting and bead appearance, and cuts spatter, while the CO2 contributes the penetration and cost advantage. Pure CO2 never disappeared, though. It remains common on structural and heavy fabrication where penetration matters more than cosmetics, and where the spatter is handled by slag-catching habit rather than by a change of gas.

The Navy, Subarc, and the Speed Factor

The industrial proving ground for shielding economics was the U.S. Navy Bureau of Ships. Its wartime shipbuilding programme pushed two developments at once: CO2-shielded semiautomatic welding for the yards, and a self-shielded flux-cored wire process the Navy called Subarc. The self-shielded wire carried its own flux and shielding ingredients inside the core, so the welder needed no gas cylinder at all. That mattered enormously in ship construction, in yards, and on any job where a gas line was impractical.

Flux-cored arc welding debuted in 1957, with self-shielded wire used on automatic equipment that greatly increased welding speeds. The welded hull itself was no longer an experiment by then, having been proven commercially as early as 1930, so the shipyards knew exactly what welding could achieve. What they lacked was throughput. CO2-shielded gun welding and high-speed self-shielded flux-cored lines solved precisely that, and wartime urgency accelerated the whole learning curve. Marine-specific practice is covered in our guide to AJ Marine welding processes, standards and safety practices.

Sheet Metal and the Auto Plants

The same economics pulled GMAW into a new industry. GMAW is used extensively by the sheet metal and automobile industries, where it is often applied as arc spot welding, replacing riveting or resistance spot welding on body panels. Arc spot welding pushes a short, high-speed weld through a stacked joint without a continuous bead, which is faster, lighter and cheaper than driving rivets or running resistance spot welders on every panel. As car bodies became larger, more complex and more corrosion-sensitive, spot welding by arc grew attractive.

By the early 1960s, MIG was no longer the process a fabricator used because they could get an inert gas. It was the process a fabricator used because nothing else welded steel as quickly or as cleanly. The story of the next chapter is how the transfer modes themselves became separate inventions.

Short Circuit, Spray, and Pulse: The Transfer Modes That Built the Process

Ask a welder how they set a MIG machine and you will usually hear wire feed speed, not volts. Ask how metal actually travels from the wire to the puddle and most shops call it “the transfer mode.” That phrase sounds like a menu option on a modern power source, but each of the four modes was a separate invention, and each one opened a door into work the previous one could not touch. The order they arrived in explains why your machine behaves the way it does.

Spray transfer: the method that came first

Spray transfer was the first metal transfer method used in GMAW. The wire melts, a droplet forms at the tip, and the force of the arc blasts it across the gap as a fine spray of small droplets. The result is essentially spatter-free, which is why spray became the natural choice for aluminium and stainless with inert shielding gas. The catch is heat. Spray needs high current and voltage, produces a large pool and a wide heat-affected zone, and reaches maximum deposition rates of about 600 mm/s (1,500 in/min). Because of that heat input it is generally used only on workpiece thickness above roughly 6.4 mm (0.25 in); anything thinner risks burnthrough and distortion.

In the early 1960s the spray-arc variation arrived, using small additions of oxygen to inert gases. The oxygen lowered the current needed to sustain spray transfer, trimming the heat and widening the range of material spray could handle.

Globular transfer: the cheap CO2 route to steel

When argon was too expensive for ordinary steel shops, the industry reached for carbon dioxide, and globular transfer was what CO2 gave them. The droplet grows larger at the tip and is transferred mechanically rather than blown across the gap. Deposition reaches up to roughly 110 mm/s (250 in/min), but the arc runs hot, the bead surface comes out rough, and spatter increases sharply. That spatter and high heat keep globular transfer limited to flat and horizontal positions. It was fast, affordable, and the mode that made steel work possible before pulse arrived.

Short circuit: the variation that took over

The short-circuit, or short-arc, variation was released in 1958 and 1959 and immediately became the most popular GMAW variation. The wire feeds down into the puddle, touches it, bridges the gap and extinguishes the arc, then surface tension pulls the droplet off and the arc restrikes. The whole cycle repeats roughly 100 times per second. The operating window is about 100 to 200 amperes at 17 to 22 volts.

Because the pool stays small and heat input stays low, short circuiting works in all positions and handles thin material well, exactly where spray and globular could not follow. Its weakness is the flip side of that low heat: on thicker material the arc extinguishes before enough energy has been deposited, producing a risk of lack of fusion and a cold-looking bead.

Pulsed spray: the modern refinement

Pulsed spray is the extension of spray transfer that the modern machine was really built around. It needs a power source pulsing between 30 and 400 times per second with a mostly-argon gas mix. The machine drops average current and heat input while keeping the instantaneous peak high enough to stay in spray, so the pool shrinks, distortion drops and welding can be done in any position. The trade-off is a lower ceiling on speed, about 85 mm/s (200 in/min).

Pulse mattered most for thin aluminium and stainless, where conventional spray burns through before you can lay a bead down, and for robotic work, where every pulse is identical and spatter is kept to a minimum. It is the standard recommendation in our aluminum MIG welding guide, and it only works on clean metal, which is why good prep matters as much as the machine settings.

The newest chapter is Cold Pulse, a pulsed-transfer technique widely credited to university research in the 1990s and 2000s and commercialised in the 2010s. It targets aluminium and stainless with minimal heat input, trimming distortion further on thin, sensitive work.

Mode	Behaviour	Speed / Window	Best For	Weakness
Spray	Fine blown spray, spatter-free	Up to ~600 mm/s (1,500 in/min)	Alu, stainless, over ~6.4 mm (0.25 in)	Burn-through, distortion, costly gas
Globular	Large droplets, mechanical transfer	Up to ~110 mm/s (250 in/min)	Carbon steel with CO2	Hot, spattery, poor bead; flat only
Short circuit	Droplet bridges gap, ~100 cycles/sec	100–200 A, 17–22 V	Thin material, all positions	Lack of fusion on thick plate
Pulsed spray	Current pulsed 30–400 Hz	Capped ~85 mm/s (200 in/min)	Thin alu, stainless, robotics	Lower max deposition rate

Remember that shielding still comes first in every mode. If the gas is wrong, none of these numbers will save the bead, and for aluminium you have to start with clean, oxide-free metal, which is covered in our guide to prepping aluminium for welding.

The Machine Evolves: Guns, Wire Feeders, and Inverter Power Sources

The clever part of 1948 was not the wire, the gas bottle, or the torch. It was the power source. Everything since has refined the same four things: a welding gun, a wire feeder, a power supply and a gas supply. A modern machine may look nothing like the first Hobart units, but the circuit is still the same circuit.

As gas metal arc welding became standard, so did the shape of the station: gun, wire feed unit, power supply, electrode wire, shielding gas. Pull the gun apart and you find the same parts a 1950s shop had.

The Anatomy of a Welding Gun

• Control switch (trigger) — the welder’s throttle; it energises the feeder and, in many machines, the power source.
• Contact tip — copper or copper-plated, carrying current to the wire. Its bore must match the wire exactly; it passes no shield gas.
• Power cable — the heavy conductor from the work lead back to the machine.
• Gas nozzle — shapes the shielding gas column around the arc. Cheapest part on the gun, most often replaced.
• Electrode conduit and liner — the coiled tube guiding wire to the gun; the liner keeps the wire straight.
• Gas hose — from regulator and flowmeter to the gun handle.

Burn-Back and Bird-Nesting

Two mechanical problems have trailed MIG since the beginning, and every redesign since answers one of them.

Burn-back happens when the wire melts back into the contact tip instead of feeding forward: the tip swells, the wire jams, the feed stalls. It is usually a voltage or gas-flow symptom rather than a broken gun: too much voltage for the feed speed, too little shielding gas, or a tip too small for the wire, or worn with a rough, oversized bore.

Bird-nesting is the wire bunching like a bird’s nest at the feeder, where it leaves the drive rolls. It comes from a feeder gripping harder than the torch is pulling, a kink or rough spot in the liner, or a mismatch between wire diameter and drive roll groove. The wire must push and pull through the same path without being crushed.

Wire Size and Feed Speed

Electrode diameters typically run from 0.7 to 2.4 mm (0.028 to 0.095 in), with wire up to 4 mm for heavy structural work. The thinnest, up to 1.14 mm (0.045 in), suit short-circuit transfer, which makes thin-gauge welding controllable. Spray transfer generally wants 0.9 mm (0.035 in) and larger.

Wire diameter	Typical pairing	Where it fits
0.7 to 1.0 mm (0.028 to 0.035 in)	Short-circuit	Thin sheet, auto body panels
1.0 to 1.14 mm (0.040 to 0.045 in)	Short-circuit or pulse	General fabrication, stainless
1.2 to 1.6 mm (0.045 to 0.063 in)	Globular or spray	Plate steel, structural work
1.6 to 2.4 mm and up (0.063 to 0.095 in)	Spray	Heavy plate, shipbuilding

Semi-automatic GMAW typically runs wire feed rates of 2 to 10 m/min (75 to 400 in/min); high-end feeders push as far as about 30 m/min (1,200 in/min) for heavy plate and spray work. Understanding the relationship between wire feed speed and the machine’s output turns a mystery box into a tool you can predict.

From Rotating Generator to Inverter

The first shop MIG machines used rotating generators: large, loud, mechanically sluggish. Silicon-controlled rectifiers (SCR) replaced them in the 1970s, quieter and more efficient but still behaving like transformers. From the 1980s onward, inverters took over: by switching current at high frequency and controlling it electronically, they could hold a steady output, manage a shorter arc, and shape the waveform. That is what made pulse welding, waveform control and synergic (preset) programs practical in ordinary shops rather than laboratory specials. A synergic program is a stored set of coordinated voltage and wire feed values picked by material and thickness.

Aluminium Hardware and Gas Delivery

Aluminium wire is soft and springy, so it hates being pushed through twenty feet of conduit. Two solutions became standard: the spool gun, which puts a small feeder at the torch, and the push-pull feeder, which drives the wire from both ends of the line. Both are covered in the practical guide to aluminium MIG welding.

Gas delivery is the least glamorous part of the station and a common failure point. The regulator and flowmeter set the volume, the hose carries it, the nozzle distributes it. Typical shop flow sits between 10 and 20 litres per minute, and a leak or blocked flowmeter turns a clean arc into a sputtering one. Gas is a consumable, so treat it like one.

Spatter builds up inside the nozzle and around the contact tip, disturbing the gas column, letting air into the arc and causing porosity. Clean tips and nozzles regularly, use anti-spatter compound (tip dip) to slow the build-up, and swap a worn crater liner rather than living with it.

MIG vs. Stick vs. TIG vs. FCAW: What the History Decided

Walk into a fabrication shop, a pipeline yard or an automotive plant and you will find more gas metal arc welding than anything else. That is not marketing spin. It is the practical outcome of everything in the previous sections. Gas metal arc welding is now the most common industrial welding process in the world, chosen for its versatility, its speed, and the relative ease of adapting it to robotic automation.

But history did not hand GMAW a clean sweep. Every competing process exists because it solved a problem MIG was bad at. Once you know which problem each one answers, choosing becomes obvious.

Stick (SMAW): the answer when there is no gas

Shielded metal arc welding was developed during the 1950s using a flux-coated consumable electrode, and it quickly became the most popular metal arc welding process. The coating decomposes into a shielding gas and slag that protects the puddle, which means the welder needs nothing but a machine, a rod holder and a stick. No cylinders, no flowmeter, no hose to leak.

That single fact is why stick still owns structural field welding and outdoor work. Wind, rain, cold, a job site with no gas supply: none of it matters. It is also forgiving of rust, mill scale and dirty steel in ways GMAW is not, because the flux is doing chemical work on the puddle. The trade-off is speed, penetration efficiency and bead appearance. Advantages of stick welding: why SMAW still wins lays out the case in practical terms, and it is worth reading before you dismiss a process your shop barely owns.

TIG (GTAW): perfected in 1941, decades before MIG

Gas tungsten arc welding took decades of development to perfect in 1941, long before MIG existed, and it remains the reference standard for thin material, appearance-critical joints, non-ferrous work and repair. The tungsten electrode does not become part of the weld, so you can add filler metal by hand, one droplet at a time, and you control exactly where the heat goes.

That heat control is the whole argument for TIG. You can weld a 0.8 mm sheet without melting it, or clad a stainless part without distorting it. The cost is speed and skill: TIG is the slowest of the four processes here and demands a steady hand and a good foot control. Where appearance and metallurgy matter more than throughput, TIG still wins. For aerospace work in particular, the standards are demanding enough that aerospace TIG welding standards and techniques are worth a closer look.

FCAW: built for the site in 1957

Flux-cored arc welding debuted in 1957 with self-shielded wire usable on automatic equipment, and it greatly increased welding speeds on heavier material. The shielding plume from the flux is dense and less likely to be blown away by wind than a light argon plume, which is exactly why FCAW dominates site and structural work.

Unlike GMAW, which needs clean gas coverage and is rarely used outdoors or in moving air, FCAW is better suited to outdoor construction. It also lays down more metal per pass and tolerates rough surface prep. The price is a slag or fuse residue that has to be chipped and ground off afterwards, and more fume, which matters in confined spaces. Aluminum flux core welding: what actually works covers the case where flux-cored wire is the right answer on aluminium.

How the history splits the decision

• Thin material and flat fabrication: short-circuit MIG wins on speed and thin-gauge control.
• Heavy structural and outdoor work in wind: self-shielded FCAW or stick, because neither depends on a gas plume surviving the breeze.
• Non-ferrous, thin, appearance-critical work: TIG or pulsed MIG, where heat control and a clean bead decide the job.
• High-volume repetitive work: robotized GMAW, which the history leaves with the widest path to automation.

Two federal rules frame every one of these choices. OSHA 1910.252 sets the general requirements for welding and cutting, including fume and gas hazards, and OSHA 1910.254 covers arc welding and cutting specifically, spelling out ventilation, fire prevention and protective equipment. Switching from a shielded gas process to self-shielded FCAW or stick changes your fume profile, so the local exhaust question has to be answered again.

No process won outright. Each one sits where its history put it, and picking the wrong tool usually means fighting the design rather than using it. If you are setting up a MIG machine from scratch, the complete MIG setup guide walks through gas, wire and transfer mode choices before you strike an arc.

From Shop Floor to Robot Cell: How MIG Became an Automation Workhorse

The switch from hand-welded assemblies to automated welding didn’t happen overnight. For a process that was born for a welder to guide by hand, gas metal arc welding (GMAW) would go on to become the backbone of the robot cell. The reason for this lies in one of its greatest advantages: it’s far easier for a machine to do than most other arc processes.

The First Step Into Automation

The origins of industrial robotics predate modern weld cells by a decade. The first commercial, digital and programmable robot, the Unimate, was built by George Devol in 1954. In 1961, it was sold to General Motors and put to work at the Inland Fisher Guide plant in Ewing Township, New Jersey, where it was used to lift pieces of hot metal from die casting machines. At the time, it wasn’t welding, but it proved that a machine could reliably perform a repetitive, dangerous task without the need for direct human interaction.

It wouldn’t be until decades later that this idea would truly mesh with welding. By the late 1970s, several large Japanese conglomerates began producing industrial robots similar to Unimation’s. This surge turned America’s early lead into a global industry, and robot welding became commonplace throughout automotive and heavy manufacturing. With factories looking to increase their production rates, the search for the perfect arc process to pair with these robots had begun.

Why GMAW Was Built For The Robot

Of all of the arc processes, GMAW was practically made for mechanization. The biggest factor for this is its self-regulating nature. Due to its constant-voltage power source, the length of the arc will correct itself. If the arc grows longer, the current rises and melts more wire to shorten it. If the arc grows too short, the current drops and melts less wire to lengthen it.

Because of this, a robotic arm doesn’t need to constantly micro-adjust its arc length like a stick welder would. Instead, it only needs to control three things: travel speed, wire feed speed, and stick-out distance. Additionally, GMAW is incredibly easy to mechanize with a separate wire feeder and a torch mounted onto a long conduit, allowing for the compact, durable setups that robot cells require.

From Spot Welds To Seam Tracking

The automotive industry would be the largest catalyst for robotic GMAW’s rise to prominence. GMAW saw heavy use within sheet metal fabrication, often utilized for arc spot welding in place of riveting or resistance spot welding. This allowed for continuous body panels to be joined at a much faster rate, saving both time and labor on the assembly line.

As the technology progressed, so did the parts being welded. What started off as simple spot welds evolved into multi-pass joints on more complex geometries. To accommodate this, factories began utilizing multi-axis fixtures and purpose-built welding cells. From here, more advanced refinements began to appear, such as offline programming to drastically cut down on teach-time, as well as laser-scanned seam tracking to allow robots to adapt to slight part variations. These advancements paved the way for modern setups, such as those used when working with more specialized jobs, including aluminum MIG welding.

A Legacy That Keeps Evolving

What was once a luxury only accessible to massive automotive plants in the 1980s has drastically changed. Today, modern collaborative and robotic MIG cells are more compact and affordable than ever, allowing smaller fabrication shops to automate jobs that would have been unattainable for them only a few decades prior.

Most impressively, this story of repurposing hasn’t ended on the factory floor. GMAW has also been used as a low-cost method of 3D printing metal objects. Open-source machine designs have been documented, and in many cases, aluminum components fabricated through this method have been able to compete with conventionally made parts on strength. It’s a perfect example of a process that refuses to stay stagnant, constantly finding new life in the technology of tomorrow.

From the Unimate lifting red hot castings, all the way to wire-fed metal 3D printers, GMAW proved to be so much more than a hand process. It was an automation workhorse, and to this day, it still reigns as one of the most prominent forces driving automated welding into the future.

Why We Call It “MIG”: Terminology, AWS Standards, and the GMAW Name

Here is a small puzzle that trips up nearly every new welder. MIG stands for metal inert gas. So how can a steel weld shielded with argon and carbon dioxide be MIG? Carbon dioxide is not inert. It is one of the most reactive gases you will ever put over a molten puddle.

The answer is that “MIG” stopped being a description and became a brand name. Commercialised in the early 1950s, the trade name travelled faster than any attempt at technical precision, and by the time the engineering community caught up, the word was welded shut in the vocabulary of every shop on earth. The accurate term exists; it is just used less in conversation than in print.

GMAW, MIG and MAG: Three Names for One Family

The American Welding Society classifies the process as GMAW, gas metal arc welding: solid-wire welding where an external shielding gas protects the molten weld metal. It is the umbrella term. Under it sit two regional habits.

• MIG, metal inert gas. The American and British shorthand, and accurate when the shielding gas is genuinely inert, such as pure argon or a helium blend. That is what you run on aluminium, stainless steel, copper and magnesium.
• MAG, metal active gas. The European convention, and the correct word when the shielding gas contains an active component such as CO2 or oxygen that reacts with the pool.

So a 75/25 argon-CO2 weld on mild steel is, strictly speaking, MAG. Most shops still say MIG and most machines are still sold with “MIG” on the panel, but a European spec sheet would say MAG. The process is identical either way; only the label changes. AWS keeps tubular flux-cored welding separate as FCAW, so the gas-shielded solid-wire family is GMAW and nothing else.

The Standards That Grew With the Process

Any process carrying load, pressure or safety risk cannot be governed by trade names. The American Welding Society, founded in 1919 as a nonprofit organisation whose mission is to advance the science, technology and application of welding and allied joining and cutting processes, has been writing the rules ever since and has authored more than 350 standards covering welding practices and procedures. The main families a GMAW welder meets:

• D1.1 / D1.1M, the Structural Welding Code, steel. The most widely cited document in fabrication work.
• D1.2 / D1.2M, structural welding code for aluminium.
• D1.5M, bridge welding for highway and railway structures.
• A3.0M, automotive welding, which replaced the older vehicle practice documents used in body-in-white and structural repair.
• A5-series, filler metal specifications defining AWS ER70S-6 wire, ER4043 aluminium wire and the rest of the consumable catalogue.
• A2-series, inspection documents covering how welds are examined and accepted.

The full index sits in AWS’s codes and standards library. Alongside AWS sits ASME, whose Boiler and Pressure Vessel Code governs anything holding pressure. Section IX qualifies welders and procedures for pressure equipment. Pipelines fall under API 1104; structural work under AWS D1.1.

The Machine on Your Bench Is Only Half the System

Here is the practical consequence. A GMAW machine is a heat source and a wire feeder, not a welding procedure. The code-compliant side of the work lives in the written procedure and the welder’s qualification, and that is where the money, the liability and the career protection sit.

A qualified procedure states the joint configuration, filler metal, gas and flow rate, wire diameter, travel speed, voltage and wire feed settings, cleaning method, shielding requirements and acceptance criteria. A welder qualification proves a particular person can perform that procedure within the code’s allowed variables, and that it holds up under test. Our walkthrough of welding certification and procedure qualification shows how the two documents work together.

The ladder of credentials is real and climbable. AWS delivers education in in-person, online instructor-led, self-paced and dual formats, and certification runs from the Certified Welder program through to Certified Welding Engineer and Certified Welding Inspector credentials. A welder who can weld and a welder who can write and sign off a procedure sit on very different paychecks.

So the naming confusion resolves cleanly. MIG is what people say, GMAW is what the standards say, MAG is what the rest of the world says when the gas is active. All three describe the same arc, the same constant-voltage trick and the same continuously fed wire, and all three sit inside a written code that says exactly what you must do to be allowed to say you did it properly.

History Repeats Itself: Modern Mistakes and Their 1940s Roots

Every bad MIG weld you have ever made has a historical twin, and the culprit is usually a deliberate 1940s cost decision.

Thin Sheet on Pure CO2: A 1953 Economy Problem

When carbon dioxide became the affordable shielding gas in 1953, it made MIG weldable for the average shop, but it inherited two liabilities. CO2 is an active gas, so it oxidises the molten pool, and the oxides on a fast-cooling steel puddle spit off the arc as spatter, while aggressive heat transfer leaves the pool hot and hard to tame on thin material.

Hence argon blends for short-circuit work on thin steel. The 75/25 and 90/10 mixes keep CO2’s burn characteristics while calming the arc, cutting spatter and giving a puddle you can steer on 18 gauge sheet. Pure CO2 still makes sense on thick structural and heavy plate, where speed and penetration outweigh the spatter.

Short-Circuit Settings on Thick Plate: The Frozen Pool

Short-circuit transfer is a heat-control bargain: the wire touches the pool, shorts the arc, dumps a slug of filler in, and the machine re-strikes. It is the gentlest way to fill a weld, which is exactly why it fails on heavy material.

Less arc energy means less heat into the plate, so the pool freezes at the leading edge, the pool behind it refreezes just as quickly, and a cold slug of metal sits where the puddle should be. The result is lack of fusion and insufficient penetration: the bead looks laid on top rather than melted into the plate.

Spray Transfer on Thin Material: The Burn-Through Special

Spray is the fastest, hottest, most productive transfer mode, and on 2 mm sheet it is an emergency: the arc burns a hole and leaves a heat-affected zone wide enough to ruin the part. Spray needs roughly 6.4 mm (0.25 in) of material to throw a stable pool into.

Skipped Aluminium Cleaning: The Oxide Nobody Saw

Aluminium welds fail in ways that look like bad technique but are really chemistry: aluminium oxide and nitride form the instant the metal meets air, and both are tenaciously stubborn, so a spool left open in a humid shop grows its own oxide film, and so does the plate you just cut.

Both surfaces must be brushed or chemically treated immediately before welding, and the wire should run through a spool gun or push-pull feeder that keeps it dry and uncoated. Fighting black specks in an aluminium bead? Start at the cleaning step: read how to prep aluminium properly for welding before you change a machine setting.

Porosity: What It Actually Is

Porosity is not a shielding failure in the way people assume. It is gas trapped in the metal, bubbles frozen into the weld: gas enters the molten pool, and if the metal solidifies before it can float out, it stays put as a pore. The faster the pool cools, the more gas gets locked in. Aluminium is especially prone, because its higher thermal conductivity pulls heat out of the pool fast, which is why porosity hits hardest on the first weld of a cold part and why a preheat or a slower, wider weave clears it.

Stick-Out: Too Long or Too Short

Stick-out is the distance between the contact tip and the work, a surprisingly powerful variable.

• Too much stick-out means the electrode melts too far from the work: a sputtering arc, shallow penetration, poor deposition, and gas that cannot blanket the fusion zone.
• Too little stick-out means the wire melts back into the contact tip: spatter builds inside the gun, the tip wears out early, and burn-back stops the wire feeding mid-bead.

For most solid wire, roughly 15 to 20 mm (5/8 to 3/4 in) is a sane target, adjusted for the gun and the joint.

Gas Flow, Wind, and Dirty Lines

Shielding gas is a consumable, and it fails three ways. Too little flow lets air in around the nozzle, so you weld a mixture of gas and atmosphere, a porosity machine. Too much flow creates nozzle turbulence that can suck air in at the edges, and a leaking hose or dirty regulator delivers gas that protects nothing.

Outdoors is the hard case, because drafts dissipate the shielding gas and drag contaminants into the weld. Flux-cored arc welding handles outdoor work better, because the flux in the wire core generates its own shielding, so wind matters far less. That is why bridges, pipelines and site fabrication still lean on self-shielded flux core and stick rather than GMAW. Welding aluminium outdoors? Read what actually works with aluminium flux core before you fight the wind with a spool gun.

Wire Size and Drive Rolls: The Feed Side

Mismatched drive rolls and wire diameter cause feed stalls and bird-nesting, the bunching of wire in the feeder. Knurled rolls suit hard-drawn wire, smooth or V-grooved rolls suit soft wire, and a knurl too aggressive for the wire will chew it. That failure drove two-roll and four-roll push-pull feeder design, and the fix is still simple: match roll type and groove size to the wire, check for a stuck or dirty liner, and keep the gun clean.

The machine has not changed in principle since 1948, so neither have its failures. Name the decision behind the problem, and the repair is usually obvious.

What a Welder Should Take From Eighty Years of MIG History

Most people who dig into the history of MIG welding come away with lists of dates and names. Neither changes how you weld on Monday morning. The sharper takeaway: MIG was not one invention but three problems solved one after another, and every machine today is still solving them.

Three problems, one trick that never changed

Shielding came first. From Davy’s arc in 1800 through carbon arcs and coated stick electrodes of the 1930s, every process fought the same enemy: air. Oxygen and nitrogen in a molten puddle mean porosity, brittle oxides and unstable arcs. MIG answered with shielding gas, affordable from 1953 when carbon dioxide arrived, then split the difference with argon-CO2 blends. Set a flow meter today and you are still adjusting a 1948 idea.

Heat control came second. Constant voltage made the whole thing work, because it let the arc regulate itself. The industry refined it further: globular for steel, short-circuiting for thin material, spray for heavy plate, pulsed spray for a small, cool, predictable pool. Every one of the four transfer modes in GMAW answers one question: how do we keep the puddle under control without making the welder fight the machine?

Feed handling came third. Contact tip, liner, drive rolls, spool gun, push-pull feeder. Each came from a specific failure, whether burn-back or bird-nesting, and each fix still works in your gun this afternoon.

Five habits worth keeping

• Match the process to the material. Thin sheet and flat fabrication want short-circuiting MIG. Heavy structural work and outdoor jobs in wind want self-shielded FCAW or stick. Appearance-critical and non-ferrous work wants TIG or pulsed MIG. Our guide to where stick welding still wins explains the splits.
• Understand why the settings work instead of copying a chart. Voltage sets arc length, arc length sets current, current sets how fast the wire melts, and wire melting rate sets how fast the puddle fills. See that loop and charts stop being magic.
• Buy the features you will actually use. Pulse for thin aluminium and stainless. Synergic programs if you repeat the same joint constantly. A spool gun or push-pull feeder if you run aluminium wire. And enough gas flow capacity, because a machine that cannot deliver it will fight you.
• Keep the gun honest. Contact tip, nozzle, liner. A worn tip causes burn-back and poor shielding, a dirty nozzle causes spatter, a rough liner causes drag and feeding problems. Fifteen minutes of maintenance saves an afternoon of chasing a wild arc.
• Treat gas as a consumable, not a detail. Cylinders get exchanged, mixes get contaminated, hoses leak, regulators freeze. Know what your machine needs and keep spare bottles on hand. Old lesson, easy one to forget.

The process itself is learnable. Many people reach reasonable proficiency with GMAW in a few weeks given structured training and practice, though even trained operators see quality swing with heat, wind, fit-up and material condition. If aluminium is on your bench, our practical guide to aluminium MIG welding covers the cleaning and gas details that trip people up most often.

If you want to go further with it

The demand for skilled welders is not a soft story. AWS workforce data puts welding professionals in the United States at roughly 771,000 as of 2025, with 320,500 more projected by 2029 and around 80,000 openings per year. More than 157,000 professionals are approaching retirement, and the median welding salary sits around $55,600, with average hourly earnings of $26.76. The pipeline is leaking faster than it fills.

Routes in vary. A welding certificate takes roughly six to eighteen months, a two-year associate degree is common, the journeyman card runs three to four years, and a four-year bachelor’s degree opens the engineering and inspection side of the trade. The AWS offers credential paths from Certified Welder and Certified Welding Educator through Certified Welding Inspector, Supervisor and Engineer, with training in person, online instructor-led, self-paced and dual formats. Our aluminium welding course overview shows what the training looks like day to day.

When MIG was the right answer, and when it was not

Did MIG replace stick and TIG? No. It joined them. Stick wins outdoors and on heavy structural steel because it needs no gas. TIG wins where the bead is the product. MIG took the work that rewarded speed, forgiveness and automation.

Was MIG always right? Only with shielding gas, clean metal, a machine that could hold the arc, and a joint that did not need a manual TIG torch’s control. Where any were missing, the older processes won.

Is MIG obsolete? No. It is still the most common industrial welding process in the world and the backbone of robotic sheet metal production.

The arc itself has not changed since Davy. Electricity still makes heat, metal still melts, the puddle still freezes, and a modern machine with waveform control and synergic programs does the same 1948 trick far more precisely. Eighty years of MIG history is a record of people solving that control problem one layer at a time, and the next one is probably still being designed. The tools are cheaper and the training more accessible than ever, and every setting on your machine is a decision somebody made for you decades ago, on purpose.

Frequently Asked Questions

Who invented MIG welding?

Modern gas metal arc welding is generally credited to 1948 work at Battelle Memorial Institute, where a constant-voltage power source developed by H. E. Kennedy was paired with a smaller-diameter continuously fed electrode. Industry histories also credit Hobart Brothers of Troy, Ohio, and engineer Dave McKee with commercialising the process and developing the continuously fed welding gun that made it practical.

What does MIG stand for, and is it the correct term?

MIG means metal inert gas, and the American Welding Society prefers the broader term GMAW (gas metal arc welding). Because a shielding blend containing carbon dioxide is not inert, most steel MIG welding is technically MAG (metal active gas), though ‘MIG’ is the everyday word almost every shop uses.

When was MIG welding invented?

The process dates to 1948, but its foundations go back to 1920, when P. O. Nobel of General Electric built a wire-feed predecessor with no shielding gas at all. The commercial breakthrough came in the early 1950s, and 1953 brought carbon dioxide as a welding atmosphere, which made MIG economical for ordinary steel fabrication.

How is MIG welding different from flux-cored arc welding?

MIG uses a solid wire and an external shielding gas, so the weld bead comes out clean but the process is vulnerable to wind. FCAW uses a hollow wire filled with flux, which can be self-shielded, making it far better suited to outdoor and structural site work, though it leaves a residue that has to be cleaned between passes.

Why did MIG welding become so popular so quickly?

Three advantages did it: continuous wire feed removed the need to add filler manually, the constant-voltage arc regulated itself so the welder only managed travel speed and gun angle, and the deposition rate was far higher than stick welding. When carbon dioxide became affordable in 1953, the process finally made economic sense on carbon steel.

Can you MIG weld aluminium?

Yes, but aluminium needs argon-based shielding, scrupulously clean metal and wire, and usually a spool gun or push-pull feeder to push soft wire reliably. For thin material, pulsed MIG is the usual answer because it keeps the pool small and the heat input low enough to avoid burn-through.

Is MIG welding suitable for outdoor use?

Generally not. Wind and moving air blow the shielding gas away and let contaminants into the pool, which causes porosity and weak welds. For outdoor structural work, self-shielded flux-cored wire or stick welding is the standard choice for exactly this reason.

Which metal transfer mode should I use?

Short-circuiting for thin sheet and general fabrication, spray for thick plate and stainless or aluminium in flat position, globular only where high deposition on heavy steel matters, and pulsed-spray when you need a small, cool pool, all-position capability or clean welds on thin non-ferrous material.

Do I need a pulse-capable MIG machine?

Only if your work justifies it. Pulse is the deciding feature for thin aluminium and stainless, for all-position work on thinner material, and for robotic applications where a small, spatter-free pool is easier to automate. For structural carbon steel with 0.035-inch wire, a standard machine does the job perfectly well.

How did the shift from generators and rectifiers to inverters change MIG welding?

Inverter sources from the 1980s onward gave far faster current response and precise control, which made pulse welding, waveform shaping and synergic programs possible on ordinary shop machines. It also reduced the size and weight of machines and improved arc control on thin material, where older constant-voltage machines were more difficult to tame.