When Did Welding Become a Thing? 4,000 Years of Metal Joining

A farmer in rural Ohio picks up a 60-year-old stick welder and runs a bead that looks exactly like the ones his grandfather laid down in 1962. Nothing about the machine feels ancient, yet the idea behind it — melting metal so two pieces become one — is older than the pyramids. So when, exactly, did welding become a thing? The honest answer is messier, and far more interesting, than a single date.

When Did Welding Become a Thing? The Short Answer First

There is no single date when welding became a thing, because welding is not one invention. It is a family of processes that arrived in waves, each waiting on its own enabling technology.

Here is the compressed timeline:

• Around 3000 BC: Egyptian and Near Eastern smiths forge weld copper, gold, and bronze by heating and hammering. This is real welding, not soldering.
• Around 1800: Humphry Davy demonstrates a sustained electric arc between carbon electrodes using a voltaic battery. Spectacular science, no welding application yet.
• 1880s: Nikolay Benardos patents carbon-arc welding, the first practical electric arc process.
• 1904–1907: Oscar Kjellberg develops the coated electrode, with patents from 1907 to 1914. This made arc welding trustworthy enough for industry.
• 1941: Russell Meredith patents TIG welding for aluminum and magnesium aircraft work.
• 1948: MIG welding arrives, feeding wire continuously to raise deposition rates.
• 1960s onward: Automation takes hold, from submerged arc to robots, orbital systems, and digital controls.

That is the skeleton. The flesh is more interesting.

“The Technique Existed” vs. “The Technique Became a Thing”

Most timelines confuse the existence of a technique with its arrival as a real-world practice. Those are different milestones, often decades apart. Think of four doors a process must walk through:

1. Demonstration. Someone proves the physics in a lab. Davy’s arc in 1800 lives here.
2. Patent. Someone claims the idea as property. Benardos in the 1880s lives here.
3. Commercial product. Equipment you can buy, with consumables, manuals, and service. Lincoln Electric’s move into arc welding equipment around 1911 lives here.
4. Codified industrial practice. Codes, qualified procedures, trained operators, inspection. The founding of the American Welding Society in 1919 lives here, along with the first edition of the ASME Boiler and Pressure Vessel Code in 1915.

When people ask when welding became a thing, they usually mean door four: the moment it stopped being an experiment and became something engineers specified, inspectors checked, and insurers accepted. For arc welding, that is roughly 1900 to 1925.

What This Article Means by “Welding”

To date anything, you have to define it. This article uses the metallurgical definition: welding is the coalescence of metals at the atomic level, achieved with heat, pressure, or both, with or without filler metal. That includes the ancient smith’s forge weld and excludes soldering and brazing, where the filler melts at a lower temperature than the base metal and the bond is more adhesive than metallurgical.

Use the loose definition, “sticking two metal things together,” and welding is older than civilization. Use the narrow one, “electric arc welding,” and it is a late nineteenth-century technology. Both answers are defensible. Only one is useful without context.

Where the Evidence Comes From

Historians work from archaeology, patents, trade catalogs, and standards. Artifacts and microstructure show Bronze Age joints were genuinely welded, not glued or riveted. Patents establish priority but not practical success. Catalogs and engineering journals show when a process reached paying customers. Codes show legitimacy: once a process enters a code, it is no longer experimental. As the Encyclopaedia Britannica entry on welding notes, the field’s history is usually divided by process rather than one chronological spine, because the technologies developed in parallel, at different speeds, in different countries.

The Three-Part Answer You Actually Need

By the end, you will have three overlapping answers instead of one.

Ancient roots. Welding as a craft begins around 3000 BC with forge welding. Smiths produced metallurgically continuous joints millennia before anyone understood why hammer blows worked. The Universal Technical Institute’s history of welding traces that lineage through ancient Egypt and the Iron Age to the village forge.

Nineteenth-century electrification. Welding as an electrical process begins with Davy’s arc in 1800, becomes patentable in the 1880s, and becomes commercially viable with the coated electrode around 1904. The Miller history of welding and ESAB’s corporate history both document that arc from curiosity to industry.

Twentieth-century industrialization. Welding as an organized engineering discipline begins with the codes, training programs, and inspection regimes of the 1910s and 1920s, then explodes into specialized processes from the 1930s through the 1980s, from TIG and MIG to submerged arc and laser.

Each story gets its own treatment ahead. If you remember one thing, remember this: welding never had a birth. It had a childhood, an adolescence, and a long, noisy adulthood. The advantages of stick welding that keep it dominant in field work today descend from the coated-electrode breakthrough of 1904, which is why the history is worth understanding.

What Actually Counts as Welding? Definitions That Change the Date

Ask ten people when welding was invented and you may get ten dates. None are necessarily wrong. The answer shifts because “welding” is not one thing. It is a label for a family of joining methods, and each definition moves the starting line.

The Metallurgical Line: Welding, Brazing, Soldering, and Fastening

At the strictest level, welding means coalescence: the atoms of two pieces of metal actually bond across the joint. Filler metal is optional. Heat, pressure, or both do the work. Once the joint cools, the base metal and weld zone behave as one continuous piece, though the microstructure around the seam has changed.

Brazing and soldering sit on the other side of a clear temperature line. In brazing, the filler melts above 450°C (840°F) but below the melting point of the base metal. Soldering uses a filler that melts below that same threshold. In both, the base metal never melts. The filler wets the surface and forms a metallurgical bond, but that is not the same as melting the parent metal. A small torch on copper pipe is usually brazing or soldering, not welding. The OSHA overview of welding, cutting, and brazing treats them as related but distinct processes for good reason.

Mechanical fastening — bolts, rivets, clamps — creates no atomic bond at all. The pieces simply cannot separate because something stronger holds them. That distinction kills two common mistakes: ancient smiths were not gluing metal together, and they were not merely soldering. When they heated iron to a bright yellow and hammered it, they forced oxide aside and pressed clean surfaces into contact until diffusion did the rest. That is a real weld. The history of welding at TWI describes forge welding as one of the oldest joining methods, and the archaeology backs it up.

Two Big Families: Fusion and Solid-State

Welding splits into two broad families. Fusion processes melt the base metal: arc, gas, resistance, laser, and electron beam all fall here. Solid-state processes skip the melting stage and rely on pressure, heat, or both to bring atoms together. Forge welding, friction welding, explosion welding, and diffusion bonding belong to this group.

The family tree explains why “welding started in the 1800s” is a bad answer. The 1800s gave us electric fusion welding. They did not give us welding. Forge welding was already thousands of years old by then. Modern laser welding is a fusion process with a very recent toolset, but it descends from the same idea as an arc: melt the interface and let it solidify. You can see that lineage in practical guides to laser welding systems, where the vocabulary of fusion, shielding, and heat input is still front and center.

The Vocabulary Problem: “Autogenous Soldering” and “Burning Together”

Part of the confusion is linguistic. “Welding” only became a standard English trade term in the late nineteenth century. Earlier writers reached for phrases like “autogenous soldering” or “burning together.” Autogenous meant joining metal without a separate filler, which sounds like welding to a modern reader but was often used for processes we would now classify differently. The word changed before the technique did.

That gap is why old patents and trade catalogs mislead. A process described as “soldering” in 1860 might be a true weld by today’s definition, while a “welding” claim in 1900 might describe a joint that was really brazed. A welding textbook history chapter makes the same point: terminology evolved unevenly across trades and countries.

Process Families and Approximate First Appearances

Process family	Approximate first appearance	What defines it
Forge / solid-state	Bronze Age, around 3000 BC	Heat and hammering create a metallurgical bond
Oxyfuel gas	Late 1800s	Acetylene discovered in 1836; practical torches came later
Carbon arc	1880s	Arc between a carbon electrode and the workpiece
Metal arc	1890s	Consumable metal electrode melts into the joint
Resistance	1885	Electric current plus pressure at the interface
Thermite	1890s	Aluminothermic reaction supplies intense heat
TIG	1941	Tungsten electrode with an inert gas shield
MIG	1948	Continuously fed wire electrode
Laser	1960s	Focused light beam melts a narrow joint

Demonstration, Patent, Product, Practice: Four Different Dates

Even after we agree on a definition, we still have to agree on what counts as the date. A laboratory demonstration proves a process can work. A patent proves someone claimed it first, not that anyone used it. A commercial product proves a company believed it could sell the idea. A codified industrial practice proves engineers trusted it enough to write it into standards, inspection rules, and training programs.

Those milestones can sit decades apart. Resistance welding was patented in 1885, but it took the sheet-metal industries years to build machines and procedures around it. TIG was patented in 1941, yet it became routine only after aircraft manufacturers and their suppliers worked out the shielding, torch design, and operator training. A history of welding from Miller Electric shows the same pattern again and again: invention arrives first, adoption lags behind.

So the definition sets the date. If welding means any metallurgical joining of metal by heat and pressure, the story starts in the Bronze Age. If it means an electric arc melting metal, the story starts in the 1880s. If it means a reliable, code-governed industrial process, the story starts in the early twentieth century. All three are legitimate answers — the trick is to say which definition you are using before you name a year.

Ancient Roots: Bronze Age Forge Welding and the First Real Welds

A charcoal hearth glows in the dark. A smith pulls a bundle of iron bars from the coals, checks the color, and brings it to the anvil. A few quick hammer blows, a sprinkle of sand, another heat, and the pieces merge into one solid bar. No arc, no shielding gas, no machine settings. Yet the joint is real metallurgical welding, and it is thousands of years older than the electric welder in your shop.

The First Real Welds: Egypt and the Near East

The earliest evidence generally accepted by historians comes from Egypt and the surrounding Near East, around 3000 BC. Metalworkers were joining copper, gold, and bronze by heating small pieces and hammering them together. The equipment was simple: a charcoal hearth, a blowpipe or early bellows, a stone or bronze anvil, and a hammer. The skill was not.

At the right temperature, clean metal surfaces can bond under pressure. Oxide and slag have to be kept out, the pieces need to touch intimately, and the hammer has to work fast enough to consolidate the joint before it cools. Get it wrong and the layers separate. Get it right and the boundary can disappear under a microscope. That is why the history of welding does not begin with electricity. It begins with fire and force.

Iron Changes Everything

The Iron Age, from roughly 1200 BC onward in the Near East, raised the difficulty. Iron does not melt and flow like bronze at the temperatures ancient furnaces could reach. Instead, smiths learned to bring iron to a near-white heat and squeeze the slag out while welding the surfaces together. This is forge welding, also called fire welding, and it became the foundation of ironwork for roughly four thousand years before arc welding appeared.

Smiths developed clever ways to build reliable pieces from imperfect material. In piled construction, they stacked and welded iron bars, drew them out, folded them, and welded again, refining the metal with every cycle. Pattern welding went further: layers of iron and steel were forge welded into a billet, twisted or stacked, welded again, and then ground and etched. The acid attacked the layers differently, revealing the swirling pattern associated with Damascus-style blades. Sword makers also used laminated cores, welding harder steel edges to tougher iron bodies so the blade could hold an edge without snapping. These were not decorative tricks. They were engineered answers to inconsistent raw materials.

The Iron Pillar of Delhi: Famous, but Not a Lost Welding Secret

No ancient iron object attracts more speculation than the Iron Pillar of Delhi. Dated to roughly 300–400 CE, it stands over 7 meters tall and weighs several tonnes. It has survived more than 1,600 years of monsoon weather with remarkably little corrosion, which has led to claims that ancient smiths possessed a lost welding method superior to modern techniques.

The metallurgy tells a more grounded story. The pillar was built up by hot working iron into a single piece, and many accounts describe forge-welded bundles or bars. What is certain is that it is not a modern fusion weld, and it is not evidence of some secret arc process. Its corrosion resistance is largely attributed to a protective phosphate-rich layer that forms on the iron surface, linked to the metal’s phosphorus content and the environment. That is chemistry and skilled ironworking, not a lost welding torch.

Village Forges and the Smith’s Standing

Roman and medieval blacksmiths turned forge welding into everyday work. They welded chain links, hinges, tools, and agricultural implements. They joined iron barrel hoops and repaired wagon tires. In villages across Europe and Asia, the local smith was a essential community figure: the person who could repair a plow before planting season, straighten a bent axle, or rebuild a broken hearth tool. The work was dirty, hot, and highly skilled, and the smith’s reputation rested on joints that held.

That repair instinct never disappeared. It survives today in specialist metal repair, from alloy wheel welding to structural fabrication. The tools changed; the goal did not.

By the time the first electric arc was demonstrated, forge welding had already shaped tools, weapons, buildings, and transport for millennia. The metals changed, too. Modern aluminum behaves nothing like bloomery iron, which is why aluminum boat welding follows its own rules. But the ancient lesson held: clean surfaces, the right heat, and controlled pressure make two pieces of metal behave like one.

The Electric Spark: Davy, Acetylene, and the Battery Problem

On a lecture bench around 1800, Humphry Davy hooked a voltaic pile to two carbon rods, touched their tips together, then drew them slightly apart. A blinding arc jumped the gap and kept burning. It was a magnificent parlor trick of science. It was not yet welding. The pile could sustain the display only while its chemicals lasted, and that power was costly and limited. Davy had revealed a phenomenon, not a manufacturing process. The moment belongs in the long history of welding, but as a starting gun for the electrical age, not the finish line.

That distinction sits at the heart of this part of the story. The electric arc eventually fathered stick, TIG, MIG, and submerged arc welding. In 1800, however, nobody had a generator, a stable electrode, or steel consistent enough to trust. The arc was a solution waiting for a problem it could actually solve.

Davy’s Arc Was a Spark, Not a Workshop Tool

The first customers for electricity were light, chemistry, and communication. Arc lamps lit streets and lighthouses. Electroplating deposited thin metal coatings. Telegraphy carried messages down copper wires. Inventors chasing those markets pushed dynamos, carbons, and circuits forward. Joining metal was nowhere near the top of the list. An arc that could bore a hole through a boiler plate looked more like damage than progress.

There was also the battery problem. A voltaic pile produced a modest, expensive current by consuming zinc and acid. Running a welding arc from one would have been like powering a factory from a suitcase of lantern batteries. Practical arc welding had to wait for dependable, affordable electricity in the 1880s, when generators and distribution finally caught up with the idea. Until then, the arc remained a laboratory spectacle with a very short leash.

The metals were not ready either. Cast iron was common but unpredictable under heat. Early steels varied from batch to batch, and high-carbon scraps could harden and crack beside a weld. A process needs repeatable materials as much as repeatable power. Nineteenth-century metallurgy was still learning to deliver both.

Acetylene Changes the Chemistry

In 1836, Humphry’s cousin Edmund Davy isolated acetylene, a gas that burned with a remarkably hot, bright flame. For decades it was a chemical curiosity. Then came the practical pieces: reliable cylinders, regulators, torch valves, and cheap calcium carbide for generating the gas. Mixed with oxygen, acetylene produced a flame hot enough to melt steel, braze joints, and cut plate. Oxyfuel gas welding became commercially viable in the late nineteenth and early twentieth centuries, giving fabricators a portable heat source that needed no electrical supply at all.

That mattered enormously in repair shops, farms, and field work. A welder could carry the fire to the job. Even today’s smaller air-acetylene torches carry that same logic: controlled flame, simple equipment, no outlet required. The oxyacetylene torch also taught metalworkers to read a puddle, control heat input, and manage distortion, lessons that transferred directly to arc welding.

The Industrial Revolution Was Pushing Hard

Meanwhile, the demand side of the equation was exploding. Railways needed locomotives, rolling stock, and track. Steam boilers needed shells and seams that would not leak or burst. Shipyards needed hulls that could survive rough water. Riveting could do the job, but it was slow, noisy, labor-intensive, and prone to leakage between overlapping plates. Every rivet hole also removed metal from the joint.

Engineers wanted stronger, tighter, lighter connections. Boiler explosions made the cost of bad joints brutally clear. That economic pressure, more than scientific curiosity, set the stage for the patent rush of the 1880s. Once generators could deliver serious current and steel became more predictable, the arc stopped being a lecture demonstration and started looking like a tool. The gap between Davy’s spark and a working weld shop was not a failure of imagination. It was a supply-chain problem: power, electrodes, metals, and trained hands all had to arrive together.

By the 1880s, the ingredients were finally in place. The next step was not a better battery. It was a patentable process that could turn an arc into money.

The Invention Decade: Patents, Electrodes, and the Birth of Arc Welding

The 1880s did not invent the idea of joining metal. It invented the paperwork. In a single decade, arc welding moved from a laboratory curiosity into a stack of patents, and that is the moment most historians point to when they say welding “became a thing” in the modern industrial sense. The technology was rough and the results were inconsistent, but the foundation was poured.

Nikolay Benardos and the Carbon Arc

Russian inventor Nikolay Benardos is credited with patenting carbon-arc welding in the 1880s. His setup was simple: a carbon electrode, an electric arc, and a workpiece. The arc melted the metal at the joint, and the operator usually fed a filler rod into the puddle by hand. No wire feeder, no gas cylinder, no digital display. Just heat, carbon, and a very steady hand.

That simplicity was also the problem. The carbon arc was brilliant for heating metal, but it was not a controlled welding system. The arc wandered, the heat spread, and the resulting weld depended almost entirely on the operator’s skill. Still, it was the first practical electric arc welding process, and it proved that electricity could do more than light streets. As TWI’s history of welding notes, these early carbon-arc experiments opened the door to everything that followed.

From Carbon to Metal: Slavianov and Coffin

The next logical step was to stop using carbon and start using the filler metal itself as the electrode. Nikolay Slavianov experimented with metal electrodes in the early 1890s, and Charles L. Coffin received a US patent in 1890 for arc welding with a bare metal electrode. The appeal was obvious: the electrode could melt into the joint and become part of the weld, eliminating the separate filler rod.

In practice, bare metal electrodes were a headache. The arc was unstable, spatter was heavy, and the molten metal was exposed to the surrounding air. Oxygen and nitrogen rushed into the weld pool, creating porosity and brittle joints. Welders today take shielding gas for granted. These pioneers had nothing but hope and a short arc gap.

Elihu Thomson and the Quiet Revolution in Resistance Welding

Not every breakthrough came from an arc. In 1885, Elihu Thomson patented electric resistance welding, a process that passed current through two overlapping pieces of metal and squeezed them together. The resistance at the contact point generated heat, and pressure completed the joint. There was no arc, no filler, and no open flame.

That principle became the backbone of sheet-metal manufacturing. Spot welding and seam welding could join thin-gauge steel quickly and repeatably, which made them a natural fit for appliances, cans, and eventually automobile bodies. While arc welding struggled with porosity, resistance welding offered speed and consistency in the factory. It was less romantic than a blue arc, but it was quietly building the modern assembly line.

Goldschmidt’s Thermite: Welding by Chemistry

Hans Goldschmidt took a different route entirely. In the 1890s, he developed the aluminothermic process, better known as thermite welding. A mixture of aluminum powder and iron oxide was ignited, producing a violent chemical reaction and a pool of molten iron. No electrical supply, no gas bottles, just a crucible and a reaction hot enough to melt steel.

The process was dramatic, and it found a perfect home in tram rails. Rail joints could be welded in place, creating a continuous running surface without the need for a power source in the street. Thermite welding later became a standard technique for railway track and heavy repairs, and it remains in use today. It was proof that welding did not have to be electrical to be industrial.

Why the Decade Mattered, and Why It Hurt

The 1880s and 1890s also saw early experiments with stud welding and flash welding, where current and pressure were used to join fasteners or butt sections of metal. These processes were not yet mature, but they showed that the underlying physics could be pushed in several directions. The patent office was filling up with possibilities.

The harsh reality, though, was that most of these processes produced welds that engineers did not trust. Bare electrodes left welds porous and brittle. Arc control was poor. Inspection was primitive. Riveting was slow, but it was understood, and structural engineers stuck with it for decades. Early arc welding survived on repair work, thin sheet, and stubborn optimism.

Process	Key figure	Approximate date	What it was actually good for
Carbon-arc welding	Nikolay Benardos	1880s	General arc heating, repairs, thin sheet, manual filler work
Bare metal-electrode arc welding	Charles L. Coffin, Nikolay Slavianov	c. 1890–1893	Simpler equipment and self-consuming filler, but porous, brittle welds
Resistance spot and seam welding	Elihu Thomson	1885	Fast, repeatable joining of thin sheet metal in factories
Thermite welding	Hans Goldschmidt	1890s	Rail joints, heavy repairs, work sites without electrical power
Stud and flash welding experiments	Various inventors	1890s	Fasteners and butt joints, still experimental and unreliable

The bare-electrode era also explains why modern electrode selection is so fussy. The lessons learned from unstable arcs and contaminated welds eventually led to coated electrodes, shielding gases, and careful procedure control. You can still see that lineage in something as ordinary as an alloy steel welding electrode, where the coating does the job that bare wire could not: protect the arc and shape the weld.

By the end of the 1890s, welding had patents, inventors, trade press coverage, and a growing list of failures. That may not sound like a revolution, but it was the necessary one. The next step was not a new idea. It was making the existing idea reliable enough to trust.

The Coated Electrode Revolution and Welding’s Industrial Breakthrough

By the early 1900s, arc welding worked. It just did not work well enough to trust. Bare metal electrodes produced a fierce, wandering arc, spattered molten metal, and left behind brittle, porous welds. Engineers looked at the results and chose rivets. The missing piece was not more electricity. It was control over the atmosphere around the arc.

That control arrived through a modest-looking idea: wrap the electrode in a mineral coating. The change turned arc welding from a workshop experiment into a construction process.

Oscar Kjellberg and the Electrode That Stopped Fighting Itself

In Sweden, Oscar Kjellberg began experimenting with covered electrodes around 1904. He patented improvements between 1907 and 1914, and the results were difficult to ignore. The covering was not merely insulation. As the arc burned, the coating broke down and did several jobs at once:

• It stabilized the arc. Easier starting and a steadier flame made consistent welds possible.
• It created shielding gas. Gases released from the coating pushed away air that would otherwise contaminate the molten metal.
• It formed a protective slag. The slag blanket slowed cooling and protected the weld from oxidation.
• It could add alloying elements. The coating helped adjust the weld metal’s composition and properties.

Those four functions explain why the coated electrode mattered so much. Earlier welders were fighting the atmosphere and the arc at the same time. Kjellberg’s approach gave them a tool that did part of the job automatically. The basic principle still sits in every stick electrode used today, which is why stick welding remains so stubbornly useful in wind, mud, and repair work.

Kjellberg founded ESAB in 1904, and the company’s own history traces its origins to those electrode experiments. The timing matters. This was not a single lab demonstration. It was a product, a company, and a manufacturing method arriving together.

America Catches Up, and War Demands Speed

Across the Atlantic, the equipment side was moving too. Lincoln Electric began manufacturing arc welding equipment around 1911, a shift documented in historical summaries of the process. Welding was no longer dependent on imported apparatus or improvised power supplies. It had a commercial supply chain.

Then came World War I and a shipbuilding emergency. Riveted hulls required skilled crews, overlapping plates, and thousands of holes drilled and filled. Arc welding promised faster assembly and, just as importantly, watertight joints. Shipyards began testing the process on real structures, not sample plates.

The rush exposed weaknesses. Electrical systems in yards were often inadequate. Weld quality depended heavily on the operator, and trained welders were scarce. Inspection was primitive by modern standards. A weld could look sound on the surface and hide slag, porosity, or incomplete fusion underneath. The history of welding is full of this pattern: a new process creates a new failure mode, and the failure mode creates a new inspection method.

Writing Welding Into Engineering Law

Two institutions gave the technology legitimacy. The first edition of the ASME Boiler and Pressure Vessel Code appeared in 1915. It was published before welding was widely codified, but it established the principle that pressure-containing equipment would be governed by written rules rather than individual shop judgment. Welding would eventually have to fit into that system.

In 1919, the American Welding Society was founded. Its mission and history reflect the moment: welding needed standards, shared terminology, and a way to transfer knowledge between industries. A process that had been learned by apprenticeship was becoming a field with published practices.

Safety lagged behind. Early arc welding exposed operators to intense light, sparks, metal fume, and electrical shock. Ventilation was often an open door. There was no modern fume regulation to require local exhaust or respiratory protection. Those protections came much later, including the federal rules that now govern welding fumes and electrical safety.

From Gut Feeling to Inspection and Training

During the 1920s and 1930s, welding changed from a craft judged by appearance to a process judged by evidence. X-ray inspection was adopted in that period and exposed internal defects that no visual check could find. A welder could no longer rely only on a steady hand and a clean bead.

Electrode classification followed. Manufacturers began grouping electrodes by coating type, position, and current, giving engineers a common language for specifying consumables. Training programs multiplied, and welders learned not just how to strike an arc but how to follow a procedure. The modern idea of a qualified welder, tested against a written standard, grew out of this era.

By the late 1930s, coated-electrode arc welding had earned a place in shipyards, pressure vessels, buildings, and repair shops. It was still noisy, dirty, and dependent on skill. But it was no longer a curiosity. It was a construction method with a supply industry, a standards body, and an inspection record behind it. That foundation made the next wave of processes, from TIG to automated welding, possible.

The Twentieth-Century Explosion: TIG, MIG, Submerged Arc, and Automation

By the 1930s, welding had proven it could hold ships and boilers together. The next five decades turned it into a toolbox. Each new method answered a specific complaint: the arc was too exposed, the plate too thick, the aluminum too fussy, the seam too long, or the run too repetitive. Welding accumulated options rather than replacing itself.

Submerged Arc: Taming the Arc for Heavy Plate

Submerged arc welding, or SAW, emerged in the 1930s for a problem that plagued heavy-steel fabricators: arcs are bright, smoky, and unstable at serious current. SAW buries the arc under a blanket of granular flux. The wire feeds continuously, the flux melts into protective slag, and the arc burns invisibly beneath the pile.

Because no one watches an exposed arc, the process runs at very high currents. That suits thick plate and long, straight seams: pressure vessels, storage tanks, structural beams, and pipe. It is not a field process: the work usually needs positioning and rotation, and the flux must be handled and recovered. But on long joints and heavy plate, few processes match its deposition rate and consistency.

TIG: A Solution for Aircraft Aluminum

The tungsten inert gas process, TIG, was patented in 1941 by Russell Meredith. Its first serious application was welding aluminum and magnesium aircraft structures, materials that punished older, dirtier arc processes. TIG separates electrode from filler: a non-consumable tungsten electrode establishes the arc, and the welder adds filler by hand when needed. An inert gas shield protects the pool.

The result is precise, clean, and controllable at low currents, which made it valuable for thin sheet, stainless, and aerospace work where contamination or distortion could ruin a part. Slower than wire-fed processes and more demanding, it never became the default for heavy construction. For aircraft skins and critical tube assemblies, the trade-off was worth it. Modern aerospace TIG welding still follows the same logic: control the heat, protect the pool, keep the joint clean.

MIG and Flux-Cored Wire: Feeding the Arc

MIG welding arrived around 1948 with a simple but powerful change: a continuously fed consumable wire electrode. Instead of stopping to change a stick every few inches, the welder keeps going. The wire melts into the joint and acts as filler while a gas shield protects the arc. Deposition rates climbed, and semiautomatic production welding took off.

That made MIG a natural fit for automotive frames, sheet-metal panels, trailers, and general fabrication. It is fast, forgiving on clean steel, and easier to learn than TIG. Aluminum is a different animal: soft wire, stubborn oxide, and feed-system quirks. Our aluminum MIG welding guide covers those details.

Flux-cored arc welding came along in the 1950s for outdoor and high-deposition work. The wire contains flux that generates shielding gas as it burns, so FCAW is less vulnerable to wind than plain MIG and useful on structural steel, bridges, and repairs. It also leaves slag to clean and is not a substitute for every solid-wire job.

Vertical Thickness, Concentrated Energy, and New Limits

Electroslag and electrogas welding were developed for vertical, thick-section work where ordinary arc welding would need many passes. Electroslag uses a molten slag bath to melt the wire and joint faces in one controlled pass; electrogas uses a gas-shielded arc with a moving dam. Both are niche today but solved a real problem: welding a thick vertical seam without turning the whole assembly.

Plasma arc welding concentrated arc energy for precision work and keyhole welds at moderate thickness. Electron beam welding arrived in the late 1950s, firing a focused electron beam in a vacuum to produce a narrow, deep weld with minimal distortion. Laser welding followed the 1960 ruby laser, offering speed, precision, and low heat input on thin sheet and automated lines. Modern laser welding systems descend from that work, though equipment and safety requirements have changed enormously.

Process	Problem it solved	Sweet spot
Submerged arc	Invisible arc at high current	Thick plate, long seams
TIG	Clean welds on reactive metals	Aerospace, thin aluminum
MIG	Continuous wire feed	Production steel, sheet metal
Flux-cored	Wind-resistant shielding	Structural steel, field repair
Electron beam / laser	Deep, narrow, low distortion	Precision, automated lines

Automation: From Unimate to Orbital Welding

The first Unimate industrial robot was installed at General Motors in 1961. It was not a welding specialist, but it proved a machine could repeat a hazardous, repetitive task. Welding robots spread through the 1970s and 1980s as automotive plants chased consistent beads, faster cycle times, and relief from dangerous positions. Fixed robots handled repeated parts, orbital systems handled pipe, automated cells took long runs.

Today robotic and orbital welding sit alongside manual processes rather than replacing them. A pipeline spread may use orbital welding for consistent root passes, while a repair crew still relies on stick in a muddy ditch. An automotive plant may run dozens of robots on one body line; a custom fabricator uses one welder and a MIG machine. Volume, access, material, and joint variation decide.

That is the era’s real lesson. Submerged arc did not make stick obsolete. TIG did not eliminate gas welding. Lasers and robots did not erase the need for a welder who can read a puddle. Each claimed a corner of the map defined by material, thickness, position, and production volume. As the history of welding shows, progress came from adding capable options, not one replacement moment.

Why the Timeline Still Matters in a Modern Shop

Walk into any fabrication shop and you are walking through a museum where the exhibits still run. Every process leaning against a wall or plugged into a socket is a fossil of a specific problem somebody once failed to solve with the tools they had. That is not sentiment. It is practical: when you choose a process, you are choosing a very old answer to a question about wind, thickness, metal, or volume.

Your Process Choice Is a History Lesson

Reach for stick on a windy bridge site and you are borrowing Oscar Kjellberg’s 1904 idea. The flux coating that stabilized his arc still does the same job today, generating shielding gas and slag that keep the atmosphere off the puddle when gas cylinders and windbreaks are impractical. That is why SMAW still wins outdoors and in repair work. TIG on thin stainless or aluminum traces straight back to Russell Meredith’s 1941 aircraft work, where thin skins needed precision and clean arcs more than speed. And MIG’s continuous wire feed, arriving around 1948, exists because someone wanted to deposit metal faster than a hand could swap rods.

Job condition	Likely process	Historical reason
Outdoor field repair, dirty steel	Stick (SMAW)	Flux-coated electrode shields the arc from wind
Thin stainless, aluminum, cosmetic work	TIG (GTAW)	Developed for thin aerospace skins in 1941
Production runs on mild steel	MIG (GMAW)	1948 wire feed raised deposition rates
Thick plate, long horizontal seams	Submerged arc	1930s answer to heavy fabrication
High-strength low-alloy structures	Low-hydrogen, controlled heat input	Learnned from hydrogen cracking failures

Codes Carry the Scars

Standards are history written into law. AWS D1.1 governs structural steel welding, ASME Section IX controls procedure and performance qualification, and API 1104 rules pipelines. Each clause exists because something cracked, leaked, or fell down. When a welder earns a certification, that paper is a chain back to those failures — proof that the lesson survived.

Materials Forced the Toolbox Open

Wrought iron was forgiving. Mild steel was still tolerant. Then high-strength low-alloy steels arrived and demanded controlled heat input to avoid cracking and losing toughness, which is why procedure limits on amperage, travel speed, and preheat are not bureaucracy. Stainless needed shielding that stopped carbon pickup. Aluminum needed cleaning discipline and AC or pulsed current, which is why prepping aluminum properly still decides whether a joint passes. Every new alloy in the last century pulled another process out of the drawer.

Settings Don’t Weld — People Do

Modern synergic MIG power sources adjust voltage automatically as wire feed speed changes and choose programs for you. Useful, and it lowers the entry barrier. It does not read the puddle. Watching the toe wet in, hearing the arc settle, catching porosity before it becomes a repair — none of that is automated. The welder’s hand is still the last control loop.

So the timeline is not trivia for a quiz. It explains why the shop floor looks the way it does, and why the right answer to “which process?” was worked out long before you arrived.

Myths, Mistakes, and How to Date Welding Correctly

Search “when was welding invented” and you get a tidy answer, usually a year, usually wrong. The story is messier, and most of the confusion comes from a handful of myths copied from site to site without anyone checking a patent number. Here is how to dismantle them, then how to judge a historical claim for yourself.

Myth One: “Welding Was Invented in 1881”

False. 1881 is roughly when practical carbon-arc welding was patented, not when welding appeared. Smiths were fire-welding iron centuries before anyone understood electricity. Even in the electrical era, the 1880s produced patents and demonstrations, not a finished industry. A patent is a legal document describing an idea; a usable process needs power supplies, electrodes, trained hands, and customers willing to trust a joint over a rivet. Blurring those stages is the single most common error in welding history.

Myth Two: “The Ancients Only Soldered”

Also false, and metallurgically illiterate. Soldering and brazing melt a filler that flows between parts at a lower temperature than the base metal. Forge welding heats the base metal itself until the surfaces coalesce under pressure. A well-made ancient iron weld is a genuine metallurgical bond, and metallography on historical blades routinely shows it. The confusion is vocabulary: the word “welding” only settled into its modern trade meaning in the late nineteenth century. Earlier writers said “burning together” or “autogenous soldering.” Different words, same physical reality.

Myth Three: “The Iron Pillar of Delhi Proves Ancient Welding Was Superior”

The Iron Pillar of Delhi, roughly 300 to 400 CE and standing about seven meters tall, is genuinely remarkable for its corrosion resistance. But that resistance comes from chemistry, particularly a phosphorus-rich slag layer that forms a protective passive film, not from a lost welding secret. It was likely assembled by forge welding overlapping iron blooms, a normal technique of the period. Calling it proof of superior ancient welding confuses a materials story with a joining story.

Myth Four: “Arc Welding Was Instantly Adopted”

Riveting dominated structural steel into the 1930s, decades after arc welding was available. Engineers had good reasons to hesitate: bare electrodes produced brittle, porous welds, inspection was primitive, and a bad rivet fails locally while a bad weld can fail a whole frame. Some early all-welded ships suffered brittle fractures, which pushed research into steel toughness, weldability, and fracture mechanics. The same skepticism greeted stick welding in plenty of shops, which is why stick welding’s stubborn advantages still get argued over today.

The Safety History Nobody Puts on the Timeline

Early welding was dangerous in ways the pioneers barely registered. Fumes, manganese exposure, hexavalent chromium from stainless, electric shock, arc flash, and fires in wooden shipyards were all part of the job. Regulation arrived late. The Occupational Safety and Health Act was signed in 1970, and welding rules were codified at 29 CFR 1910.252. Today OSHA limits manganese fume exposure to a 5 mg/m³ ceiling, while Cr(VI) in welding fume is regulated at 5 µg/m³ as an eight-hour time-weighted average. The broader OSHA welding and cutting pages spell out ventilation, eye protection, and fire watch requirements that earlier generations simply did not have.

How to Date a Welding Process Yourself

You do not need a research library. You need a method that forces evidence.

1. Define the process precisely. “Arc welding” covers carbon arc, bare metal arc, and shielded metal arc. They have different dates.
2. Separate the four milestones. Laboratory demonstration, patent filing, commercial product, and codified industrial practice can sit decades apart.
3. Find the primary document. A patent number, an excavation report, or a dated trade journal beats a blog post every time.
4. Check the trade press of the era. Engineering, The Welding Journal, and similar publications show what shops actually used, not just what inventors filed.
5. Look for the standard. A process becomes industrial when a code writes rules for it, which is why bodies like the American Welding Society matter to the timeline.
6. Cross-check two independent sources. If both trace back to the same undated list, you have found one source, not two.

A Checklist for Any “First” Claim

• Does the source name a patent number, inventor, and filing date?
• Does it distinguish a demonstration from a product?
• Does it cite an excavation report for archaeological claims?
• Does it mention adoption problems, or only triumphs?
• Is the publication dated, and is the author identifiable?
• Does it repeat the same phrasing as a dozen other pages?

Apply that filter and something useful appears: the honest sources are the ones willing to say “we don’t know exactly.” For comparison, watch how process histories handle the same gaps, from carbon arc gouging’s evolution in air arc welding and carbon arc gouging to the electrode developments covered elsewhere in this article. Dating welding correctly is less about memorizing years than about asking where a number came from.

Conclusion: Welding Didn’t Start Once — It Kept Starting

So when did welding become a thing? After four thousand years of evidence, the honest answer is that it became a thing four separate times, each answering a different question. Once as a craft. Once as an electrical curiosity that turned practical. Once as an industrial system with rules written around it. And once as an automated, software-driven technology one operator can supervise. That kills the tidy single-date answer search results love. Welding wasn’t invented in a flash of insight. It accumulated.

The Four Moments Welding Became Real

1. Around 3000 BC — the craft. Smiths in Egypt and the Near East heated and hammered copper, gold and bronze into true metallurgical joints. No patents, no standards — just fire, charcoal and a hammer.
2. The 1880s — the electrical process. Carbon arcs, metal electrodes, resistance welding and thermite all arrived inside about fifteen years. The history of welding turns genuinely modern here.
3. 1900s to 1920s — the industrial system. The coated electrode, the founding of ESAB, the first ASME code and the American Welding Society turned a workshop trick into engineering law. If you still reach for stick because it works in wind and rain, you are using a descendant of Oscar Kjellberg’s 1904 idea.
4. 1960s onward — automation and control. Unimate at General Motors in 1961, robots spreading through the 1970s and 1980s, and digital monitoring today.

Every Leap Came From a Failure

Look closely and the through-line is failure, not genius. Bare electrodes produced brittle, contaminated welds, so flux coatings arrived. Hydrogen cracking ruined high-strength joints, so low-hydrogen electrodes and baking ovens followed. Riveting was too slow for a wartime shipyard, so arc welding got its chance. Thin aluminum aircraft skins resisted every existing process, so Russell Meredith gave us TIG in 1941 — the same lineage behind modern aerospace TIG work. Welding has always advanced by fixing whatever just broke.

A One-Paragraph Timeline to Carry With You

Forge welding appears around 3000 BC and stays the smith’s main tool for millennia. Davy observes a sustained arc in 1800; acetylene arrives in 1836. Practical carbon-arc patents land in the 1880s, metal electrodes and thermite in the 1890s. Coated electrodes arrive around 1904, World War I forces adoption, and the AWS forms in 1919. Submerged arc, TIG, MIG and flux-cored wire follow between the 1930s and 1950s, then electroslag, electron beam and laser. Robotics takes over from the 1960s; hybrid and additive processes build on principles demonstrated in the 1960s and 1970s.

Where to Go Next, Depending on Why You Asked

• If you’re here for history: read a patent rather than a listicle. Start with Coffin’s 1890 filing, then compare it to a modern electrode datasheet.
• If you’re a hobbyist: try a real forge weld. One fire weld teaches you more about coalescence than ten articles.
• If you’re career-focused: pick a process family and go deep — aluminum MIG is a strong place to start, since it rewards the wire-feed logic invented in 1948.
• If you inspect or engineer: read the standard, not the summary. Codes carry the scars of every failure described in this article.

What Comes Next

Laser-arc hybrid welding, wire-arc additive manufacturing, collaborative robots and real-time monitoring all extend ideas demonstrated decades ago, now affordable thanks to better sensors and software. Automation is one of fabrication’s fastest-growing categories, and the Bureau of Labor Statistics still projects steady demand for welders, cutters, solderers and brazers across hundreds of thousands of US jobs. Machines will do more of the depositing. Someone still has to read the puddle, judge the fit-up and know why the procedure says what it says.

That’s the real answer. Welding became a thing when a smith first hammered two hot pieces into one, and it has been becoming a thing ever since. Every arc you strike carries four thousand years of accumulated fixes, all aimed at the same stubborn goal: making two pieces of metal behave like one.

Frequently Asked Questions

When did welding first begin?

The earliest evidence of forge welding dates to roughly 3000 BC in ancient Egypt, where metalworkers heated and hammered metals until they bonded. Electric arc welding, the process most people mean today, was patented in the 1880s.

Who invented welding?

No single person did. Ancient smiths developed forge welding over millennia, Nikolay Benardos patented carbon-arc welding in the 1880s, Charles L. Coffin patented a bare metal electrode process in 1890, and Oscar Kjellberg developed the coated electrode around 1904.

Was welding used before electricity?

Yes. Forge welding was the dominant method for iron for thousands of years, and it is still used by blacksmiths today. Oxyfuel gas welding also predates practical arc welding.

Why is 1881 sometimes called the birth of welding?

That date is tied to early carbon-arc welding patents and demonstrations, which made electric welding practical for the first time. It is a milestone for arc welding, not the beginning of welding itself.

When did TIG and MIG welding appear?

TIG welding was patented in 1941 by Russell Meredith, largely for aircraft aluminum. MIG welding arrived around 1948 with a continuously fed wire electrode, which raised deposition rates for production work.

What did the coated electrode change?

The flux coating stabilized the arc, produced shielding gas, formed slag over the bead, and could add alloying elements. That made welds far less porous and brittle than bare-electrode work, which is why arc welding became an industrial method.

Is the Iron Pillar of Delhi actually welded?

It is famous for its corrosion resistance and ancient ironmaking, but it is not a clear example of large-scale welding. Its durability comes mainly from a phosphate-rich protective layer, not from a lost welding technique.

When did welding become regulated for safety?

Welding safety rules became enforceable in the United States with the Occupational Safety and Health Act of 1970, followed by specific requirements at 29 CFR 1910.252 covering ventilation, eye protection, fire prevention and fume exposure.

When did welding robots start being used?

The first Unimate robot was installed at General Motors in 1961, and welding robots spread through automotive plants in the 1970s and 1980s. Today robotic and collaborative welding are routine in high-volume fabrication.