Heliarc Welding

Write it down: aluminium melts at 660 °C, but its oxide skin melts at about 2072 °C. The skin survives the metal beneath it.

Sand a corner of an aluminium coupon bright, then leave it. Within minutes it dulls again — the oxide reforms in air almost instantly.

Record argon and helium in group 18: full valence shells, so no chemistry with a molten weld pool at any temperature it will reach.

Argon is 1.78 g/L and helium 0.18 g/L against air at 1.29. Argon sinks onto the weld; helium floats off it.

Cut and sand three coupons bright and weigh each to 0.01 g. The comparison depends on them starting the same.

Heat coupon A to dull red with the torch and hold 60 seconds. Cool on firebrick. This is the unprotected control.

Set coupon B in a shallow tray, flood it with argon at a low steady flow, then heat identically. Keep the gas flowing while it cools.

Do the same with helium on coupon C. Note how much harder it is to keep a blanket — helium is lighter than air and leaves at once.

Brush and weigh each coupon. Rank the mass change: open air worst, argon best, helium in between because the blanket kept drifting off.

The argon-shielded coupon should be visibly brighter with far less scale. That colour difference is what a TIG welder is looking for on a finished bead.

Note the three reasons: it is denser so it stays put, it is cheaper as a by-product of air separation, and it needs less flow for the same coverage.

Tungsten melts at 3422 °C, so it carries the arc without being consumed. Record the consequence: heat and filler become independent controls for the first time.

An electric arc emits ultraviolet light strong enough to burn the cornea in seconds. The injury is called photokeratitis, or “arc eye”. It is not felt at the time — the pain arrives six to twelve hours later, typically in the middle of the night, and feels like sand under the eyelids. It normally heals in a day or two, but repeated exposure is cumulative.

Nobody in the room may look at an arc unshielded, including bystanders. Ordinary safety glasses do not protect against it and neither does a quick glance. Use an auto-darkening helmet, screen the work so passers-by cannot see it, and warn anyone nearby before striking. The same ultraviolet burns exposed skin like strong sunburn, so cover arms and neck.

TIG produces more ultraviolet than most processes, not less — the clean, spatter-free arc is not shielded by smoke or slag the way a stick-welding arc partly is. Welders report skin burns from TIG at distances that surprise them. Cover everything.

Inert does not mean safe to breathe. Argon is denser than air, displaces oxygen and pools in low spaces such as pits, tanks and inspection trenches. It gives no warning at all — there is no smell and no urge to breathe, because the body detects carbon dioxide, not missing oxygen. Confined-space asphyxiation from shielding gas kills people every year. Ventilate, and never work in a pit with gas flowing.

Use a lanthanated or ceriated tungsten electrode, never a thoriated one. Thoriated tungsten — the traditional red-banded type — contains thorium dioxide, which is radioactive. Handling a rod is a minor exposure; grinding it makes inhalable radioactive dust and contaminates the grinding wheel. Lanthanated and ceriated electrodes perform as well or better and carry no radiological hazard, which is why this blueprint specifies them and why much of industry has already switched.

US 2,274,631, “Welding torch”, Russell Meredith, filed 4 January 1941, granted 24 February 1942. Meredith was at Northrop Aircraft, and the problem was specific: magnesium airframes. Northrop trade-marked the process Heliarc and licensed it to Linde Air Products. The formal name is gas tungsten arc welding, GTAW; everyone says TIG.

Helium first, argon almost immediately after. The name records the gas that was used at the start and was quickly replaced. Argon is roughly ten times denser than helium, so it blankets the pool instead of drifting off it, and it is far cheaper because it is a by-product of the air separation that industry already ran for oxygen. Helium survives for special cases — it carries more heat into thick aluminium and copper, so it is used in mixtures where penetration matters more than cost. The trade name outlived the technique.

The oxide problem, and the elegant trick that solved it. Argon alone does not remove the aluminium oxide already on the surface, only prevent more. The answer was alternating current: on the half-cycle when the electrode is positive, ions bombard the work and physically blast the oxide skin off — cathodic cleaning — while the other half-cycle puts the heat back into the metal. AC TIG cleans and welds in the same arc, and that is why aluminium is welded on AC and steel on DC.

Why it is still the quality benchmark. TIG separates the heat from the filler completely: the arc is set by the machine and the foot pedal, and the filler rod is fed by the other hand at whatever rate the pool wants. Nothing else gives that much control. It is slow and it demands genuine skill, so it does not compete with MIG or submerged arc on production work — but stainless pipework, aerospace assemblies, titanium and thin-section joints where the weld must be right rather than fast are still TIG, eighty-four years on.

The thread that runs from 1886 to 1942 ends here. Resistance welding avoided the problem by never opening a pool. Every process after it opened one and then argued about how to protect it — with carbon, with slag, with a coating, with a blanket, with a flame's reducing zone. Meredith's answer was to stop arguing and surround the pool with an element that refuses to react with anything. Fifty-six years from Thomson's clamped wires to a noble gas, and the question was the same the whole way.