Shielding Gas Guide: Argon, Helium and Backside Protection for Micro Welding

Shielding gas is the atmosphere the weld solidifies inside. On reactive metals like titanium the difference between a good gas plan and a poor one is visible before the part cools. This guide covers what shielding gas prevents, which gas fits which material, how to design coverage on titanium, and how to read defects back to a gas failure.

What shielding gas actually does

The four things it prevents

Oxidation forms scale and oxide films that flake off in service. Nitridation dissolves nitrogen into austenitic stainless and titanium and precipitates brittle nitride phases. Discoloration on stainless reads as a mechanical-properties reject on AWS D17.1-graded work. Porosity forms when dissolved gases escape during solidification.

On pulse arc: it also protects the tungsten electrode

Tungsten in an oxygen-containing atmosphere at arc temperatures oxidizes immediately. The oxide spits into the puddle as inclusions and reshapes the tip into a bulb that shifts arc length shot to shot. The electrode selection guide covers tip life once coverage is in order.

On laser: coverage matters even without an arc

A common misconception is that fiber and pulsed-Nd:YAG welding at 1064 nm is self-shielded because no arc draws oxygen down. Titanium, reactive stainless and refractory alloys oxidize on the cooling pool as readily under a laser as under an arc.

On resistance spot: usually not needed, but the exceptions matter

The sandwich geometry keeps the fusion zone enclosed between electrode faces and surrounding sheet. Exceptions: projection welds on reactive alloys where the joint pops above the sheet, and hermetic-seal work where surface oxidation on the sealed edge would compromise leak-tightness.

The common gases and their properties

Argon

The baseline for almost everything. Inert, monatomic, and heavier than air, so it settles into and around the puddle. Under ISO 14175 it sits in the I1 (inert) subgroup. At 99.99% purity it is the sensible starting point on any TIG or pulse arc setup.

Helium

Also inert but much lighter than air, requiring higher flow rates against buoyancy. Thermal conductivity roughly ten times argon's produces a hotter, wider arc column and deeper penetration at the same current (ISO 14175 group I2). Used on thicker sections and — critically — on copper.

Argon-helium mixes (75/25, 50/50 and others)

Blending trades cost against penetration. A 75/25 mix keeps most of the coverage stability of pure argon while gaining measurable penetration; a 50/50 mix shifts further toward helium's heat delivery. Ratios are material-and-thickness decisions.

Forming gas (95% N₂ / 5% H₂ or Ar/H₂ blends)

A reducing atmosphere for backside purge on stainless. The 5% hydrogen fraction actively strips oxide during cool-down, which is why a purged root pass comes out silver where an argon-purged one would already show straw. Argon-hydrogen blends are safer in confined spaces.

Nitrogen

Rarely correct on stainless because austenitic grades absorb it and precipitate brittle chromium nitrides on cooling. Occasionally used on copper for cost reasons; the tradeoff is losing the flexibility to run one shielding across multiple materials.

Why CO₂ doesn't belong in micro welding

CO₂ is a MIG/MAG shielding gas that dissociates in the arc into oxygen and carbon monoxide — an oxidizing atmosphere that fluxes the puddle on heavy structural steel but shows up as a defect on any micro-scale precision weld.

Material-by-material picks

The table is a starting-point selector. Thickness, joint geometry, and cosmetic requirements can shift the pick from the class shown.

The table selects the class of gas; a sample weld on the actual joint geometry confirms the setup.

Titanium: trailing cup and backside protection

Why titanium is unforgiving

Titanium starts absorbing oxygen and nitrogen into its lattice above roughly 300°C. Above 500°C the reaction accelerates and produces alpha case — a hard, brittle oxygen-enriched surface layer that must be mechanically removed on any load-bearing part. Alpha case is parent metal chemistry changed, not a removable coating.

Reading the weld color as a pass/fail gauge

The AWS D17.1 aerospace fusion welding standard sets acceptance criteria by weld tint on titanium. Silver (fully shielded) and straw (light oxide, acceptable on many specs) are the operator's targets. Bronze is borderline; purple, blue, gray and white are rejectable, escalating from tint drift to a condemned part.

Trailing cup design

The torch's standard gas cup shields perhaps 15 to 20 mm of hot metal at the arc. Once the torch moves past, the metal behind it is still above 300°C and absorbing atmosphere. A trailing cup — a physically larger, mesh-lined shield — extends the argon envelope 30 to 100 mm behind the arc, letting the metal cool below the reaction threshold while still shielded. Cup geometry is joint-specific.

Backside purge

Any open-root weld needs a purge on the underside, or atmosphere attacks from behind while shielding defends the front. On tube and pipe work this means a purge dam sealing off a section around the weld and flooding it with argon. Purge starts before the arc — long enough to displace air below 100 ppm oxygen — and holds through solidification.

Post-flow: the number people cut too short

Post-flow is the argon that continues after the arc extinguishes. On carbon steel two seconds is enough; on stainless five to eight; on titanium 15 to 30 depending on mass. Post-flow duration is coupled to the current fall profile covered in the welding waveforms guide.

Flow rate, nozzle design, and flow regime

Flow rate isn't linear with coverage

Turning the flowmeter higher does not linearly improve shielding. Above a threshold set by nozzle geometry the gas column becomes turbulent and draws air in from the surrounding room. A weld shielded at 30 L/min may be actively worse than the same weld at 12 L/min. Industry-generic starting ranges for TIG-class micro work sit around 8-15 L/min through a #6 to #8 nozzle.

Nozzle diameter and standoff distance

Nozzle diameter sets the shielded footprint at the work; standoff determines how much of that footprint has degraded into turbulence by the time it arrives. Doubling standoff roughly doubles air entrainment, which is why small increases in torch height often show up as sudden coverage failures.

Laminar vs turbulent flow

Laminar flow is the shielding-gas ideal: a smooth cylindrical column that displaces air cleanly. Turbulent flow mixes room air into the shield through eddies. A gas lens — a stack of fine screens inside the nozzle — breaks the gas into a low-velocity uniform column, extending laminar behavior to lower flow rates and longer standoffs. On titanium or reactive-stainless work it is standard equipment.

Gas purity and how it shows up on the weld

Purity grades explained

Industrial argon is typically 99.99% pure (N4.0). Ultra-high-purity is 99.999% (N5.0). The extra nine is really about which contaminants are held to what ppm level: water vapor, oxygen, hydrocarbons. On titanium and reactive stainless the difference between N4.0 and N5.0 is visible on the finished weld color.

Regulator, hose and connector quality

A ppm-controlled cylinder feeding through a leaky regulator or cracked hose delivers whatever atmosphere leaks in. Every fitting is a chance for room air to permeate the gas path. Braided stainless hose, metal-to-metal fittings and a matched stage regulator hold cylinder purity to the torch; rubber hose and quick-connects do not.

When N5.0 is worth the extra cost

Titanium in any grade, thin-wall reactive stainless, refractory metals, and any part where cosmetic weld color is a rejection criterion all justify N5.0. Cylinder cost delta is small against scrap on a rejected aerospace part. On 304 stainless outside cosmetic specs, N4.0 is enough.

Moisture is the silent killer

Water in the gas line dissociates in the arc into oxygen and hydrogen. The oxygen oxidizes the puddle; the hydrogen dissolves in and comes out as porosity on solidification. Common sources are unpurged new hoses, damp cylinder valves after storage, and low-temperature condensate at fittings.

Defect diagnosis by weld appearance

Blue, straw or gray coloration on stainless or titanium

Each color maps to a temperature-oxygen exposure envelope. On titanium the AWS sequence tracks how far into oxidation the metal went before shielding recovered. On stainless, straw is light oxide, blue is chromium oxide, gray is full oxide breakdown.

Porosity and pinholes

Dissolved gas escaping during solidification. Root causes: turbulent flow entraining air, coverage loss during cool-down from short post-flow, moisture in the supply line dissociating in the arc. Diagnosis starts with flow rate and post-flow.

Tungsten inclusions on pulse arc

Tungsten transferred to the puddle when coverage collapsed or when the tungsten touched the pool. Shows as bright metallic specks in the bead and high-density inclusions on radiography. See the electrode selection guide for tip preparation once shielding is ruled out.

Crater cracks with a discolored center

Post-flow cut too short. The arc terminates, the pool solidifies, and the last-to-freeze center is exposed to atmosphere. Extend post-flow so the crown drops below the reaction threshold before shielding ends.

Shielding needs by process

Laser micro welding

Autogenous, but reactive metals still need coverage. Typical setups use a coaxial or side-jet argon nozzle at 5-15 L/min. On titanium and reactive stainless a trailing cup extends coverage over the cooling weld. See laser micro welders for the machine-side context.

Pulse arc / micro TIG

Always requires gas. Coverage protects puddle and tungsten simultaneously; the two failure modes appear together when flow drops. Post-flow settings are longer than on heavy-scale TIG because thinner tungsten overheats without coverage during shutdown. See pulse arc / micro TIG welders for the process context.

Resistance spot welding

Generally no shielding is used. The joint sits enclosed between electrode faces and surrounding sheet, so atmosphere never reaches the fusion zone. Exceptions are projection welds on titanium and hermetic-seal work where surface oxidation on the sealed edge would compromise leak-tightness. See resistance spot welders for where gas does enter the picture.

How gas coverage interacts with pulse timing

Post-flow duration is tied to how quickly the metal cools after the arc drops out, set by pulse-fall and hold-time in the waveform. Shortening the downslope shortens the shielding window the metal needs. The welding waveforms guide covers the timing side in more depth.

Frequently asked questions

What's the difference between 99.99% and 99.999% argon?

99.99% (N4.0) is industrial-grade; 99.999% (N5.0) is ultra-high-purity with contaminant limits (water vapor, oxygen, hydrocarbons) about ten times tighter. On carbon steel or 304 stainless the difference is invisible. On titanium and any part where AWS D17.1 tint criteria apply, the extra nine shows up on the weld color and justifies the incremental cost.

Do I really need a trailing cup for titanium?

On CP titanium below 1 mm the standard gas cup can be enough with a careful post-flow. On thicker sections, Ti-6Al-4V, and any aerospace-graded work, a trailing cup is standard practice. Without one the failure mode is progressive discoloration behind the arc.

Can I use nitrogen for stainless steel welding?

Not on austenitic stainless (300 series) intended for structural or corrosion service. Nitrogen dissolves into the matrix and precipitates as brittle chromium nitrides at grain boundaries. Some duplex and specialty alloys are designed to accept controlled nitrogen with an Ar/N₂ mix; those are the exceptions.

How long should post-flow be?

Long enough for the crown of the weld to drop below the reaction threshold of the base metal. Carbon steel two to three seconds; stainless five to eight; titanium 15 to 30 depending on mass. Setup basics like this are collected on the FAQ.

Why does my weld turn blue even with gas on?

Blue on stainless is chromium oxide, meaning the surface reached roughly 400°C in the presence of oxygen. If gas was flowing, oxygen still reached the weld — through turbulent flow, post-flow cut short, a moisture-contaminated line, or a leaking regulator. Diagnostic order: flow rate, post-flow duration, gas-path integrity.