Stainless Steel Pipe Welding Procedures: A Complete Guide
Table of Contents
The welder had done hundreds of carbon steel joints, so a 2-inch 316L spool looked like an easy afternoon. Two weeks later the line failed a dye penetrant check at the root, the inside of the pipe was black with oxide, and nobody could explain why a weld that looked perfect on the outside was already rusting. Stainless steel pipe does not forgive the habits that carbon steel rewards — and that is exactly what a proper welding procedure is built to manage.
Why Stainless Steel Pipe Welding Is a Different Discipline
A one-inch 316L spool comes out of the shop looking like jewelry: even bead, no undercut, crown height right where the drawing says. Six weeks later, a walkdown finds rust-colored bleed at a weld toe, and dye penetrant flags an indication at the root. The welder did nothing different from the hundreds of carbon steel joints he built before — that is exactly the problem. Stainless rewards a different set of habits, and a procedure exists to make those habits mandatory instead of optional.
So what is a stainless steel pipe welding procedure? It is the written, tested, repeatable answer to one question: how do we make this joint perform for its design life, not just until the inspection? That answer lives in three connected documents.
- Weld Procedure Specification (WPS): the recipe card at the pipe. Process, base metals, filler, joint design, amperage and voltage ranges, travel speed, purge and shielding gas, interpass limits, cleaning steps, and inspection hold points.
- Procedure Qualification Record (PQR): the evidence behind the recipe. A test coupon welded to a draft procedure and then destroyed in mechanical testing — typically bend and tensile tests, plus macro examination.
- Welder Performance Qualification (WPQ): proof that a named individual can deposit sound metal following a qualified WPS. It is the welder’s test, not the procedure’s.
The chain runs one way. Test data produces the PQR, the PQR supports the WPS, and the WPS supports each welder’s WPQ. Break a link and the joint is unqualified on paper even if it looks fine. The codes that invoke this system include ASME B31.3 for process piping, while AWS D10.4 covers austenitic chromium-nickel pipe and tubing specifically.
The Metallurgy That Writes the Rules
Stainless steel is an iron alloy with at least about 10.5% chromium, and common grades like 304 carry roughly 17–19%. That chromium forms the passive film — a microscopic chromium-oxide layer that rebuilds itself in the presence of oxygen. Scratch it and it heals. Starve it of oxygen, cover it in oil, or bake it black with heat tint, and it cannot. Nearly every rule below traces back to protecting that film.
The families behave very differently under the arc:
- Austenitic (304/304L, 316/316L, 321, 347): the workhorse piping grades. Tough, weldable, not hardenable by heat treatment.
- Duplex and super duplex: roughly half ferrite, half austenite. Strong and chloride-resistant, but sensitive to heat input and cooling rate.
- Ferritic and martensitic: magnetic, higher strength, less common in process pipe and fussier about cracking and toughness.
- Superaustenitic (6Mo): heavy molybdenum for severe chloride and acid service, often welded with nickel-based fillers.
Then there is the physics. Austenitic stainless conducts heat at about one-third the rate of carbon steel, so heat stacks up at the joint instead of spreading away. That gives you a hotter weld pool and a wider heat-affected zone. Thermal expansion runs roughly 1.5 times that of carbon steel, which is why a stainless spool moves, pulls, and warps in ways a carbon steel spool of the same size never will.
Finally, the sensitization window, which shapes more stainless procedure rules than any other single fact. Hold austenitic stainless between roughly 800 and 1500°F (427–816°C) — exactly where a slow, hot, multi-pass weld sits — and chromium carbides precipitate along the grain boundaries. That draws chromium out of the metal beside those boundaries, leaving chromium-depleted zones that cannot form a passive film. The result is intergranular corrosion: a weld that passes every mechanical test and then fails from the inside. Low-carbon L grades, stabilized grades, heat input limits, and interpass caps all exist to keep the joint out of that window or limit the time spent in it.
Quick Note: A weld can pass visual, bend, and radiographic inspection and still corrode in service. Paper compliance proves the procedure was followed. It does not prove the passive film survived — root oxidation and embedded iron are invisible on a radiograph.
Where This Guide Goes Next
The rest of this article follows the order a real procedure package gets built. We start with codes and qualification — how ASME Section IX and AWS D10.4 turn intent into documents that survive an audit. From there: alloy and filler selection, joint prep and contamination control, purging and shielding, process choice, heat input and distortion, cleaning and passivation, inspection and troubleshooting, and finally safety and cost. To see how differently heat behaves in a metal that sheds it fast, our aluminum MIG welding guide makes the stainless problem obvious by comparison.
Read in order if you are building a procedure from scratch. Jump to purging and heat input if you already have a qualified WPS and keep getting tint or root failures. Either way, the theme holds: stainless pipe is not carbon steel with a shinier surface. It is a different discipline with a documented method behind every good joint.
Codes, Standards, and Qualification: Building a Procedure That Survives an Audit
An auditor rarely starts by looking at the weld. They start with the paperwork: which code governs the line, which WPS was used, whether that WPS is backed by a PQR, and whether the welder’s name sits on a valid WPQ. Break any link in that chain and a sound weld becomes a nonconformance.
The Document Hierarchy in Plain Terms
The fabrication code sets the rules for the piping system. ASME B31.3 Process Piping covers most refinery, chemical, and process work; B31.1 covers power piping and boiler external systems. Both then hand off the “how do you prove a welder and a procedure are capable” question to ASME Boiler and Pressure Vessel Code Section IX. Section IX organizes procedure qualification requirements in the QW-300 series and welder performance qualification in the QW-350 series.
Section IX is deliberately generic. For austenitic stainless it leaves gaps that AWS D10.4, the recommended practice for welding austenitic chromium-nickel piping and tubing, fills in with practical guidance on purging, heat input, and technique. AWS D10.10 covers local heating, and AWS D18.1 addresses sanitary austenitic tube and pipe. On boiler and power work, shops often pair B31.1 with experience from outfits like Alamo Welding & Boiler Works.
PQR, WPS, WPQ: Who Supports Whom
A PQR is the evidence. You weld a test coupon, record actual parameters, and mechanically test it – bend, tensile, sometimes macro. A WPS is the instruction written from that evidence, with ranges rather than the single values used on the coupon. A WPQ is the record proving a named welder deposited an acceptable test weld using that process. The chain runs one way: test data supports the procedure, the procedure governs the welder.
Essential variables on the PQR include base metal P-number and qualified thickness range, filler F-number and A-number, position, backing, purge and shielding gas, preheat and interpass limits, heat input, and post-weld heat treatment. Change any of these beyond the qualified range and you need a new PQR – not a note in the margin.
Writing a WPS a Welder Can Actually Use
Most bad WPS documents are copied from a template and full of parameters nobody at the pipe can follow. A useful one reads like a jobcard, not a textbook. That means:
- Parameter ranges with a nominal setting, not just minimum and maximum amps.
- A purge oxygen limit in ppm, plus where and when to measure it.
- An interpass ceiling the welder can check with a temp stick or contact thermometer.
- Cleaning steps spelled out: dedicated stainless brush, solvent wipe, no chlorinated products.
- Hold points where work stops for inspection – root before fill, tint check before cleaning.
Position deserves its own line. If the WPS covers 6G, the welder needs to know whether downhill progression is permitted; if it does not say, the answer is no. Our all position welding rod guide is a useful sanity check on how position changes technique.
Quick Note: Qualified is not the same as good. A PQR can pass every bend test while the coupon still carries heavy heat tint and a slightly oxidized root. Section IX tests mechanical soundness, not corrosion resistance. Your procedure has to be stricter than the code minimum if the line will see chlorides, high purity service, or cyclic wetting.
WPQ and the Six-Month Rule
Performance qualification is where the paper meets the person. Under ASME Section IX continuity requirements, a welder’s qualification generally lapses after six months without welding with that process. Most shops track this with a simple log, but the log only works if it is honest – a supervisor signing off on work the welder never did is the kind of shortcut that surfaces in a lawsuit, not an audit. Renewal is straightforward: a new test coupon, retested and recorded, and the welder is back in scope.
The practical takeaway is to treat the whole package as living documentation. When a procedure fails in the field, the fix starts in the PQR, not at the grinder.
Choosing the Right Stainless Grade and Filler Metal
Grade and filler selection is not a lookup exercise. Two pipes can carry the same UNS number and still need different consumables because one operates at ambient temperature in a clean plant and the other runs hot, wet, or full of chlorides. The practical decision starts with the service environment, then moves to the base metal, the welding process, and finally the filler. AWS D10.4 is a useful reference for matching austenitic pipe and tubing to suitable filler metals and joint details, but the final choice still belongs in the qualified WPS.
Start With the L-Grade Logic
The “L” in 304L and 316L means low carbon, typically held to about 0.03% maximum. That matters because carbon is the element that combines with chromium at elevated temperature to form chromium carbides at the grain boundaries. Those boundaries then become chromium-depleted and lose their corrosion resistance. Lower carbon leaves less fuel for that reaction during welding and in service.
For most process piping, 304L or 316L is therefore the sensible starting point rather than the standard-carbon version. The small strength penalty is usually irrelevant, while the corrosion margin is not.
When Stabilized Grades and 316L Earn Their Cost
Grades 321 and 347 take a different approach. Titanium in 321 and niobium in 347 combine with carbon before chromium can, which protects the passive film during high-temperature service. These grades earn their premium when the pipe will run hot enough to sit in or cycle through the sensitization range, such as heat-treating equipment, exhaust systems, and some refinery service.
Grade 316L is the workhorse for chlorides and mild chemical exposure because its 2–3% molybdenum improves resistance to pitting and crevice attack. It is not a universal answer for seawater or concentrated chloride service, but it is a meaningful step up from 304L.
Duplex and Super Duplex Need Balance, Not Just Chemistry
Duplex stainless steels are designed to solidify and cool with roughly 50/50 ferrite and austenite. That balance gives them high strength and good chloride resistance, but the weld is where it can go wrong. Too much heat or too slow a cooling rate can promote excessive ferrite and sigma phase; too little heat can leave the weld ferrite-heavy and less tough.
Nitrogen additions to shielding and backing gas help restore austenite in the weld metal and heat-affected zone. Ferrite acceptance windows vary by project specification: NORSOK M-630 commonly specifies 35–65% ferrite in duplex weld metal, while EN ISO 15156-3 permits 30–70% in the welded zone. Measure rather than assume, and always defer to the governing WPS.
Superaustenitic 6Mo Grades
Six-molybdenum superaustenitic grades push pitting resistance further for aggressive chemical, marine, and flue-gas service. They are also fussier to weld. Matching fillers are not always available or practical, so high-alloy nickel-based consumables such as 625-type fillers are often specified to preserve corrosion resistance despite dilution.
Filler Metal Pairing Table
| Base metal | Common filler | Why |
|---|---|---|
| 304L | ER308L | Standard matching chemistry with low carbon |
| 316L | ER316L | Matching molybdenum for chloride resistance |
| 321 or 347 | ER347 | Niobium-stabilized weld metal for hot service |
| Carbon steel to stainless | ER309L | Higher alloy content tolerates dilution |
| 2205 duplex | ER2209 | Balanced duplex weld chemistry |
| Super duplex | ER2594 | Higher alloy matching for severe service |
Over-Alloying, Dilution, and Consumable Control
Matching chemistry is not always the right answer. Dilution from the base metal lowers the alloy content of the root, especially in a thin-wall joint, so a slightly over-alloyed filler can preserve corrosion resistance. Dissimilar joints, high-restraint joints, and hot service may also justify a higher-alloy consumable. The same logic applies when welding highly alloyed austenitic or nickel-rich materials; the principles in this guide to Alloy 20 welding procedures transfer well.
Quick Note: A filler that matches the pipe certificate is not automatically the best filler for the joint. Consider dilution, service temperature, restraint, and the welding process before accepting the default.
Consumable storage matters as much as selection. Covered stainless electrodes pick up moisture, which brings hydrogen and porosity risk, so keep them in holding ovens and follow the manufacturer’s baking schedule. GTAW filler should stay clean, dry, and identified. Verify incoming pipe, fittings, and filler with positive material identification; a mixed bundle or mislabeled box is an expensive way to ruin an otherwise good procedure.
Joint Preparation, Fit-Up, and Contamination Control
Most stainless pipe failures are decided before the arc ever starts. A beautiful cap means nothing if the bevel was ragged, the inside of the pipe was coated in cutting fluid, or a carbon steel grinding disc left free iron buried in the land. Preparation is where stainless rewards patience and punishes shortcuts.
Bevel Geometry: The Root Starts Here
A typical pipe bevel is 37.5 degrees per side, which produces a 75-degree included angle. The root face usually runs about 1.5–2 mm, with a 2–3 mm root gap. Those numbers are a starting point, not a law. Wall thickness, process, position, and whether you are using a consumable insert all change the details. The governing procedure should state them clearly.
Machine the bevel whenever possible. A machined land is square, smooth, and consistent around the circumference, which gives the root pass a predictable surface to fuse into. Grinding works, but it is easy to dish the land, taper it, or leave a burr that opens the gap on one side. A smooth, uniform land is what lets the root wet in evenly instead of chasing thin and thick spots around the joint.
Follow the qualified joint detail and the applicable fabrication standard. For austenitic piping, AWS D10.4 is a useful reference alongside the project code.
Cleanliness Is Not Cosmetic
Clean the inside and outside of the pipe before fit-up. Remove oil, grease, cutting fluid, rust, paint, and marker residue from the bevel and at least 25 mm back from the joint. Solvent residue burns into porosity and can contaminate the root; marking pens can leave traces that interfere with the weld and later inspection.
Never use chlorinated brake or parts cleaners. When exposed to welding heat and arc ultraviolet radiation, chlorinated solvents can decompose into phosgene gas. That is not a theoretical concern. The OSHA welding safety guidance warns about degreasing with chlorinated solvents before welding. Use a dedicated stainless-safe cleaner, let the surface dry, and ventilate the area.
Keep a separate set of stainless tools. Carbon steel brushes, grinding wheels, cutting discs, and even dirty work surfaces transfer free iron to the stainless surface. That embedded iron becomes a pitting initiation site, and it will show up as rust spots weeks after the job is gone. Use stainless-only wire brushes, dedicated abrasives, and clean tables. The same discipline applies to other alloys; our guide to aluminum cleaner for welding covers how alloy-specific contamination control works in practice.
Protect prepared ends between operations. Cap them, keep them off the floor, and avoid stacking them against carbon steel. A clean bevel that sits in a dusty rack overnight is no longer clean.
Fit-Up and Tacks
Fit-up accuracy matters for manual and orbital work alike. Check root gap, root face, alignment, and hi-lo around the full circumference. A high-low mismatch creates an uneven root, changes purge flow, and makes penetration unpredictable. For orbital welds, the arc sees the same gap everywhere, so a variation of a millimetre can be the difference between a clean root and a burn-through.
Tacks are part of the finished weld. They must be fully fused, cleaned, and free of porosity, slag, or oxide. Feather the ends so the root pass ties in smoothly. Bridge tacks hold the gap and are removed before welding; removable tacks can leave a crater or an arc strike that becomes a corrosion site. Never assume a tack will be buried and forgotten.
Quick Note:
A Practical Pre-Weld Checklist
Run this check at the pipe, not from memory in the shop office:
- Bevel angle, root face, and root gap match the WPS; the land is smooth and square.
- Inside and outside are clean, with no oil, grease, cutting fluid, or marker residue.
- Only stainless-dedicated brushes, discs, and wheels have touched the joint.
- Prepared ends are capped or protected and have not contacted carbon steel.
- Fit-up is checked around the full circumference for gap, alignment, and hi-lo.
- Tacks are fully fused, clean, and feathered; bridge tacks are marked for removal.
- Purge dams are placed and sealed, and the oxygen target is confirmed.
- The WPS is at the joint, with parameters and inspection hold points understood.
None of this is glamorous work. It is, however, the cheapest insurance a stainless pipe job can buy. Get the preparation right and the weld has a chance. Skip it, and even a perfect-looking cap is only hiding a problem.
Purging and Shielding: The Single Biggest Lever on Weld Quality
On stainless pipe, the weld you can see is only half the job. The root, hidden inside, decides whether the joint survives. Purging and shielding control that hidden side. Done right, the root stays bright and the passive film rebuilds. Done wrong, a weld that looks flawless outside can corrode from the inside before the line is commissioned.
A perfect cap cannot compensate for a black root, and no amount of pickling outside will reach damage on the inside.
What the Purge Actually Does
Purge gas, almost always argon, displaces oxygen and moisture from inside the pipe while the root is molten. Hot stainless steel has a strong appetite for oxygen. If oxygen is present, heat produces a thick oxide scale instead of the thin, protective chromium-oxide passive film the alloy depends on. That scale locks up chromium at the surface and leaves a chromium-depleted layer underneath, exactly where pitting and crevice corrosion begin.
The root also cannot be reached after welding. Once the joint closes, you cannot brush, grind, or pickle the inside without cutting the pipe out. Any oxide left there becomes a crevice that traps process fluid: a hiding place for bacteria in sanitary and pharmaceutical lines, a corrosion initiation site in chemical service. There is no cosmetic fix later.
Purge Methods and Their Trade-Offs
- Argon dams: Inflatable or foam dams isolate the joint and hold a local argon pocket. They use less gas and suit small spools, but must seal reliably and be removed afterward.
- Water-soluble film dams: These dissolve and flush away after welding. Convenient in sanitary work, but residue is a contamination risk if the pipe is not rinsed thoroughly.
- Inflatable plugs: Reusable and fast for repetitive work, though heat damage, slippage, and retrieval failures can ruin the purge or leave material behind.
- Full-pipe flooding: The whole length fills with argon. Excellent atmosphere control on small, open assemblies, but high gas consumption and slow to reach a stable reading.
Whatever the method, the chamber must be sealed well enough to hold a stable atmosphere. Poor fit-up, open tack gaps, and missing dams destroy that stability before welding even starts.
Measure Oxygen, Not Good Intentions
Use a calibrated oxygen analyzer and sample at the vent, not the gas inlet. A regulator gauge tells you pressure, not purity. Let the reading stabilize after the chamber has been exchanged several times. Purge volume changes as dams move, as geometry changes, and as the joint heats, so recheck before the arc starts and again after any interruption.
Inspection guidance for ASME Code piping commonly calls for measuring purge gas oxygen at the vent and confirming it is below 2% before welding begins, as part of the requirements tied to ASME B31.3 process piping. That 2% figure is a floor for general work, not a target for quality work. For good root color, aim below 100 to 200 ppm. Sanitary and high-purity tube welding often targets 50 ppm or lower.
Quick Note: A flowmeter setting is not a purge measurement. Only an oxygen analyzer at the vent tells you what the root will actually see.
Reading Heat Tint
Heat tint is a visible record of oxygen exposure, running from pale straw through brown and blue to black. Thicker oxide means greater chromium depletion and lower corrosion resistance. Silver or very light straw is the goal. Brown and dark blue mean the purge was marginal. Black means it failed. Color does not tell you how deep the damage goes, but it tells you where to look.
| Tint Color | Likely Condition | Action |
|---|---|---|
| Silver or pale straw | Good atmosphere control; passive film intact | Standard cleaning |
| Light brown | Marginal purge; some chromium depletion | Investigate purge; pickle if corrosion-critical |
| Dark blue | Significant oxide and chromium depletion | Remove tint; correct purge before next joint |
| Black | Heavy oxide; corrosion resistance badly compromised | Cut out or remove to sound metal and re-weld |
The Shielding Side
On the torch side, more gas is not better. Excessive flow creates turbulence at the cup and can entrain air into the shield, producing porosity and tint even with a good regulator setting. Use a gas lens for wider, more stable coverage, and a trailing shield on wide beads and hot fill passes where metal has left the cup but is still too hot for air. Set flow within the qualified range and verify it at the cup. A dry, stable supply matters as much as torch setup, and a proper welding gas supply guide can help you avoid contaminated or inconsistent cylinders.
Purging is not glamorous, and it is easy to rush when the crew is under pressure. But it is the single biggest lever on stainless weld quality. Get the reading, hold the reading, and the root will show it.
Choosing the Process: GTAW, GMAW, FCAW, SMAW, SAW, and Orbital
Most stainless pipe welding arguments in the shop aren’t really about skill. They’re about process selection. Put the wrong process on a schedule and the crew spends the rest of the job fighting symptoms: oxidation, distortion, slag traps, or a root nobody can inspect properly. Get it right and half the defects disappear before anyone picks up a grinder.
GTAW: The Default Root Process
Gas tungsten arc welding is the standard root process for stainless steel pipe because it separates heat input from filler addition. The arc melts the joint; you add wire when you want to. That independence is what lets a welder hold a tight, consistent root with full penetration and controlled reinforcement instead of dumping metal into a gap and hoping.
Run it on direct current electrode negative (DCEN). That polarity puts roughly two-thirds of the arc energy into the work, which gives deeper penetration and less risk of burning the tungsten. Reverse polarity shows up as a wandering arc, a melted tungsten, and a contaminated root.
For tungsten, 2% thoriated is the traditional choice: stable arc, good current carrying capacity, long life. Rare-earth alternatives ground to a point are increasingly common in shops moving away from radioactive electrodes. Match the tungsten diameter to the amperage, keep the grind lines running lengthwise, and change the point the moment it gets contaminated.
Cup and lens setup matter more than most welders admit. A standard collet body works, but a gas lens produces smoother, more directional flow and lets you use a larger cup at the same flow rate — a real advantage on stainless, where the shielding blanket is the whole game. Larger cups help on thicker-wall fill passes; a smaller cup gives access in tight roots and small-bore work.
Where the root must be guaranteed — open-butt joints, no backing, critical service — consumable inserts solve the problem. An EB-type insert sits in the bevel and melts as the arc travels, supplying predictable filler and consistent internal reinforcement. That reliability is exactly why orbital and high-purity work leans on them.
Quick Note: The root pass decides whether the joint passes inspection. Every process below can make a good fill. Few of them make a root as reliably as GTAW with a controlled purge.
GMAW Variants: Speed With a Trade-Off
Gas metal arc welding is faster, but stainless rewards patience more than speed. Transfer mode changes everything:
- Short-circuit transfer: low heat, good for thin wall and root passes, but prone to lack of fusion if travel speed and settings drift.
- Spray transfer: hot, deep, efficient — ideal for fill on thicker wall in flat or horizontal positions, unforgiving out of position.
- Pulsed GMAW: the compromise that actually works overhead and vertical, giving spray-like penetration at lower average current.
Solid wire runs with argon-rich shielding (often argon with 1–3% oxygen or a helium mix) and demands the same purge discipline as GTAW. Flux-cored stainless wire trades cleanliness for deposition rate and tolerance of rust and draft, but leaves slag that must come off. If you’re coming from another alloy and wondering why transfer modes feel so foreign, the logic in our aluminum MIG welding guide translates surprisingly well — the same threats (turbulence, contaminated wire, wrong flow) wreck both.
FCAW, SMAW, and SAW
Flux-cored arc welding earns its place on heavier wall and field work where wind and fit-up punish solid wire. The trade-off is real: slag removal between every pass is non-negotiable, because trapped slag becomes a linear indication, and fume generation runs higher than GTAW.
Shielded metal arc welding with covered stainless electrodes still wins in maintenance and restricted access — no gas line, no torch to drag into a tight corner. The catch is moisture. Stainless electrodes pick up hydrogen from humid air, so keep them in a holding oven at the manufacturer’s recommended temperature and use portable quivers. If SMAW is part of your toolbox, the fundamentals in why SMAW still wins apply here too.
Submerged arc welding belongs in the shop on large-diameter, heavy-wall pipe: high deposition, deep penetration, clean surface. Watch the flux, because recycled flux can contaminate the deposit, and carbon pickup from the wrong flux or from carbon steel contact is a genuine risk on stainless.
Orbital: Repeatability Over Speed
Orbital welding enclosed in a controlled chamber is widely used for sanitary and high-purity stainless tube, where repeatability and documented parameters matter more than speed. The arc rotates around a fixed workpiece, the chamber holds the purge, and every joint is logged — pharmaceutical and semiconductor lines essentially require it. Its limits are real: fixturing cost, limited wall-thickness flexibility, and the fact that a programmed parameter set will happily produce a bad weld if fit-up or purge was wrong.
| Process | Typical Use | Position | Main Risk |
|---|---|---|---|
| GTAW | Roots, thin wall, critical joints | All | Heat tint from poor purge |
| GMAW (short-circuit / pulsed) | Fill passes, thin-wall production | All (pulsed) | Lack of fusion |
| FCAW | Heavy wall, field work | Most | Slag inclusions, fume |
| SMAW | Maintenance, restricted access | All | Moisture pickup, slag |
| SAW | Large-diameter heavy wall, shop | Flat, horizontal | Carbon pickup, flux contamination |
| Orbital | Sanitary, pharma, semiconductor tube | Fixed rotation | Bad fit-up or purge ruins a logged weld |
Pick the process for the joint, not for the shop’s habit. GTAW for the root, a faster process for the fill, and orbital where the paperwork matters as much as the bead — then let the WPS lock the parameters down.
Heat Input, Interpass Temperature, and Distortion Control
A stainless weld can look flawless and still be ruined by the heat that made it. Amperage alone tells you little. What matters is how much energy the arc puts into each millimetre of joint, how hot the pipe gets between passes, and how the shrinking weld pulls the spool out of shape. Control those three and the procedure stays repeatable.
Heat Input: The Number That Actually Matters
Heat input is measured in kilojoules per millimetre, or kJ/mm:
Heat input = (voltage × amperage × 60) ÷ (travel speed in mm/min × 1000)
The 60 converts minutes to seconds, and the 1000 converts joules to kilojoules. Take 12 volts, 110 amps and a travel speed of 80 mm per minute:
- 12 × 110 × 60 = 79,200
- 80 × 1000 = 80,000
- 79,200 ÷ 80,000 = 0.99 kJ/mm
Time travel speed with a stopwatch over a marked length of pipe instead of guessing, and read the actual arc voltage rather than the panel setting. On short passes, arc-on time and total time differ enough to change the real number. The same voltage-current-travel-speed relationship governs other arc processes too; our aluminum MIG welding guide shows how it behaves on a very different material.
Too Hot and Too Cold Both Cause Damage
Excess heat widens the heat-affected zone, increases distortion, and drives more chromium carbide precipitation in unstabilized austenitic grades. In duplex and super duplex it encourages sigma phase and coarse ferrite, upsetting the balanced ferrite-austenite ratio. Too little heat is no safe compromise: it causes lack of fusion at the sidewalls and root, convex beads that tie in poorly, and excessive ferrite in duplex welds.
Low thermal conductivity makes this trickier than carbon steel. Heat does not spread away quickly, so the arc concentrates energy near the joint and builds steep gradients. A setting that feels mild can still overheat a thin-wall spool.
Interpass Temperature: The Limit Most Shops Ignore
Interpass temperature is the temperature of the joint just before the next pass starts. Under ASME construction, guidance generally limits interpass temperature for austenitic stainless and high-nickel alloys to 350 F (177 C), measured about 3 inches from the joint, as reflected in ASME B31.3. Duplex is commonly held to about 300 F (150 C), with heat input kept in the 0.5-1.5 kJ/mm range. These are typical windows, not universal law. The governing WPS wins.
Check with a calibrated contact thermometer or temperature crayon. Infrared guns can misread shiny stainless. Let the pipe cool naturally between passes unless the qualified procedure permits assisted cooling.
Quick Note: Interpass temperature is not preheat. Austenitic stainless normally needs no preheat, and adding it only makes the interpass limit harder to hold. Measure the metal, not the air around it.
Stringers, Weaves, and Bead Width
Stringer beads are the default for stainless pipe. A slight weave helps sidewall fusion, but wide, lazy weaving is discouraged, and many procedures cap bead width at roughly three times the electrode or wire diameter. Wide weaves put more heat into the joint, widen the HAZ, risk poor ferrite balance in duplex, trap slag and oxide, and spread residual stress over a broader area. Keep the arc on the leading edge of the puddle and let the bead stay narrow.
Distortion: Design It Out, Do Not Chase It
Stainless expands roughly 1.5 times as much as carbon steel and conducts heat poorly, so a spool can move surprisingly fast. Use back-step welding, skip sequence, and balanced welding around the circumference. Place passes symmetrically so shrinkage pulls evenly instead of stacking stress on one side.
Fixtures and strongbacks hold dimensions but add restraint and cracking risk, so use them with care. Sound fit-up and tack sequence prevent more distortion than any straightening operation. Straightening a warped stainless spool is usually a losing game: the material work-hardens, springs back, and may end up with stresses no inspection can fully map. Prevent the movement instead of fighting it.
Post-Weld Heat Treatment: Usually Not the Answer
Austenitic stainless is generally not stress relieved after welding. Furnace stress relieving is avoided because exposure to the 800-1500 F (427-816 C) range can cause sensitization, leaving chromium-depleted grain boundaries that are far less corrosion resistant. Solution annealing restores the microstructure, but it needs high temperature and rapid quenching, is often impractical for a fabricated spool, and may introduce fresh distortion.
Heat treatment is sometimes required for specific service conditions or heavily formed parts, but it should come from the governing specification and a qualified procedure, not from a hope that it will fix a bad weld. The dependable route is still controlled heat input, disciplined interpass limits, and a sequence that lets the pipe cool and shrink predictably.
Post-Weld Cleaning, Pickling, Passivation, and Heat Tint Removal
Here is the correction that settles a lot of arguments in the shop: passivation does not remove heat tint, and pickling is not just a fancy word for passivation. They are two different jobs done by two different chemistries. Passivation dissolves stray free iron from the surface and encourages the chromium-oxide film to rebuild. It only works on a surface that is already clean. Spread passivating acid over a straw-brown or blue weld and the tint will be sitting there afterward, untouched, because those oxides are far too thick and stable for a passivating bath to break down. Pickling is the aggressive step that actually strips the oxide layer, along with a thin skin of base metal beneath it.
Quick Note: Pickling removes. Passivation restores. If tint is present, you pickle or grind it off first. Only then does passivation make sense.
Mechanical Tint Removal: Fast, but Easy to Get Wrong
Stainless wire brushing, grinding, and abrasive blasting all remove tint. The caveat is that they also smear metal across the surface, and that smeared layer is often chromium-depleted and can drag iron contamination into the joint face. Brushes, discs, and blasting media must be dedicated to stainless and never used on carbon steel. Aluminum oxide, garnet, or glass bead are acceptable; carbon steel shot is not. Grinding should be done with a fresh wheel at moderate pressure, and brushing with a stainless-only cup brush, ideally one that has never touched a dissimilar alloy.
Chemical Pickling and the Reality of HF Burns
Pickling pastes, sprays, and gels are typically built on nitric-hydrofluoric acid blends. They work fast and reach places a grinder cannot, which is why they are standard on pipe interiors and complex geometries. They are also genuinely dangerous. Hydrofluoric acid burns deep, go straight for nerve and bone, and are famous for not hurting immediately, so a worker can be seriously injured before feeling anything at all.
That means a face shield over splash goggles, acid-resistant gloves and apron, forced ventilation, and no work in a pit or vessel without supplied air. Spent pickle liquor is hazardous waste and needs neutralization and proper disposal, never a floor drain. HF fumes will also etch nearby polished stainless, so keep the rest of the spool covered.
Passivation per ASTM A380 and ASTM A967
ASTM A380 covers cleaning, descaling, pickling, and passivation of stainless steel, including the classic free-iron detection methods: water wetting, high-humidity exposure, and the copper sulfate test. ASTM A967 defines the passivation treatments themselves, including several nitric acid formulations, citric acid options, and electrochemical methods, verified with tests such as copper sulfate and ferricyanide-nitric acid spot testing. In practice, citric is easier and safer to handle and dispose of; nitric remains the stronger choice where the specification demands it. High-alloy grades each bring their own quirks worth checking before you dunk them, and Alloy 20 welding procedures are a good example of how much the details shift from one family to the next.
Electropolishing: Removal and Enhancement in One Step
Electropolishing reverses the polarity and dissolves surface metal anodically in an acid electrolyte. It levels micro-peaks, enriches chromium at the surface, and produces a thicker, more uniform passive film than passivation alone. It is the standard for sanitary, pharmaceutical, and semiconductor piping, though the tank or localized wand setup costs more than a bucket of citric acid.
The Sequence That Actually Works
- Finish the weld and let it cool naturally. No water quenching on austenitic stainless.
- Remove slag, spatter, and any arc strikes mechanically.
- Remove heat tint by grinding, brushing, blasting, or pickling.
- Pickle where the specification or service demands full oxide removal.
- Rinse thoroughly. Residual acid is its own corrosion cell.
- Passivate and rinse again.
- Verify with the specified test, then protect the surface until turnover.
Cosmetic Versus Corrosion-Critical
On a dry, indoor, non-corrosive line, a light straw tint blended out with a stainless brush may be perfectly acceptable. On chloride service, sanitary piping, or anything handling high-purity water, tint is a corrosion initiation site and a crevice where bacteria settle, and removal is not negotiable. Be honest about which one you are dealing with, and write the requirement into the WPS rather than leaving it to whoever holds the grinder.
Note: A beautifully pickled and passivated weld cannot rescue a black, oxidized root. Surface cleaning fixes the outside of the joint. Only purge control fixes the inside.
Inspection, Testing, and Defect Troubleshooting
A stainless pipe weld is inspected long before the first NDT technician arrives. The procedure sets hold points, the welder controls the variables, and each test answers a different question. Under ASME B31.3, examination requirements and acceptance criteria follow the fluid service category, so a utility water line and a high-purity chemical line do not get the same scope. That is why “we always do 10% RT” is not a procedure; it is a habit.
Visual Examination: The Cheapest Test You Cannot Skip
Visual inspection catches more stainless problems than any other method, provided the inspector knows what “acceptable” looks like.
- Bead profile: uniform, without high crowns, deep valleys, or abrupt stops.
- Undercut and overlap: both are stress risers; overlap also hides lack of fusion.
- Root concavity: excessive suck-back thins the joint from the inside.
- Tint color: silver or pale straw suggests good shielding; dark blue or black means the root was unprotected.
- Spatter and arc strikes: both damage the passive film and become pitting sites.
Inside the pipe, a borescope is often the only way to see the root. Do it before the line is closed.
Surface and Volumetric NDT
Penetrant testing finds surface-breaking flaws such as cracks, porosity, and incomplete fusion at the cap. On stainless and nickel alloys, penetrant materials must be low in halogens; residual chlorides can trigger chloride stress corrosion cracking. Removal must be complete, because trapped penetrant is its own contaminant.
Radiography finds volumetric defects: porosity, slag, tungsten, and internal lack of fusion. It is strong on thickness changes and weak on tight planar cracks oriented badly to the beam. Ultrasonic testing, especially phased array, handles heavy-wall joints and planar flaws better, but it demands a qualified technique and a skilled operator.
Finally, hydrostatic or pneumatic testing proves the system, not the weld quality alone. Follow the governing code for test pressure, temperature, and safety distance. Pneumatic testing stores far more energy and needs a written safety review.
Verifying Metallurgy: Ferrite and PMI
Duplex welds need a ferrite check with a ferrite scope. Readings scatter with surface finish, curvature, and probe placement, so procedures normally specify the number of readings, the averaging method, and the locations. Do not chase a single number.
Positive material identification confirms that filler and base metal are what the paperwork says. A mixed-up ER308L on a 316L line is a corrosion failure waiting for commissioning. The same discipline applies to high-nickel work; our Alloy 20 welding guide covers those consumable checks.
Qualification Coupons: Bend Tests and Macros
Procedure qualification is verified by destructive testing. Bend tests expose ductility and fusion at the root and cap; macro examination reveals penetration, fusion, and weld profile. These tests judge the procedure, not production welds, but they predict what production will do.
Troubleshooting the Common Defects
| Defect | Likely cause | Corrective action |
|---|---|---|
| Porosity | Low or turbulent gas flow, draft, moisture, dirty joint | Check flow with a meter, shield the area, dry consumables, clean to bright metal |
| Lack of fusion | Too little heat, wrong travel angle, poor joint access | Adjust WPS within range, correct torch angle, improve bevel or gap |
| Suck-back or root concavity | Excess purge pressure, wide gap, too much heat | Reduce purge flow, tighten fit-up, lower amperage |
| Tungsten inclusion | Electrode dipping or contact | Re-grind tungsten, reset arc length, replace contaminated section |
| Cracking | Restraint, high sulfur, sensitization, sigma phase | Reduce restraint, verify filler, control heat input and interpass |
| Heat tint | Oxygen in purge or shielding | Improve dams, measure oxygen, increase gas coverage |
| Arc strikes | Poor grounding, careless striking | Grind and blend, relocate ground, retrain technique |
Quick Note: Most stainless defects are process-control failures, not welder skill failures. Fix the purge, the fit-up, the gas flow, and the heat input, and the welder usually looks a lot better.
Safety, Common Mistakes, Cost Realities, and Final Recommendations
Everything in this guide comes down to one trade: time spent controlling the process up front against money spent fixing it later. Safety comes first in that trade, because the hazards in stainless pipe welding are quiet, cumulative, and easy to dismiss on a busy shift.
Hexavalent Chromium: The Hazard You Cannot See
Stainless steels typically contain 12–30% chromium, and the same alloying element that builds the passive film also produces hexavalent chromium in welding fume. It is invisible, odorless, and a confirmed human carcinogen. OSHA’s hexavalent chromium standard for general industry (1910.1026) sets a permissible exposure limit of 5 micrograms per cubic meter as an 8-hour time-weighted average, with an action level of 2.5 micrograms per cubic meter. Cross that action level and you trigger exposure monitoring, medical surveillance provisions, and training obligations. This matters in a welding shop more than almost anywhere else: welders represent nearly half of the employees covered by OSHA’s hexavalent chromium standards.
Engineering controls come first — source extraction at the arc, local exhaust ventilation, movable fume arms, and dilution ventilation as backup. NIOSH’s engineering control database is a sensible starting point when you lay out a booth or a pipe rack. Where ventilation cannot hold exposures down, respiratory protection and designated regulated areas close the gap. OSHA’s welding safety fact sheet is worth posting next to the WPS binder.
Quick Note: Fume generation climbs with current, wire feed, and the chromium content of the consumable. A welder running 200 amps on 316L in a corner with no extraction is the classic exposure case — and the one nobody notices until air monitoring results come back.
Shielding Gases, UV Radiation, and Pickling Acid
Argon and argon blends are asphyxiants, not irritants. Argon is heavier than air, so it pools in trenches, pits, and vessel bottoms long after the cylinder valve is closed. OSHA defines an oxygen-deficient atmosphere as anything below 19.5% oxygen by volume, and an unvented purge can reach that in a surprisingly small volume. Confined space entry means atmospheric testing before entry, continuous monitoring while inside, an attendant, and a rescue plan. Your gas supply and cylinder handling guide covers storage and handling in more detail.
UV from the arc is the other daily exposure: arc eye, skin burn, and reflected radiation off bright stainless that catches people working nearby without a screen. Hot work permits, fire watch, and screens belong in the procedure, not just the site rules. During pickling, hydrofluoric acid is the serious one — HF burns deeply, attacks bone, and gives little immediate pain signal. Proper PPE, ventilation, calcium gluconate on hand, and compliant waste handling are non-negotiable, and ASTM A380 gives you the cleaning and passivation framework to build around.
The Mistakes That Cost the Most
- No purge verification. “It looked fine” is not an oxygen reading.
- Carbon steel tooling. One contaminated brush seeds a hundred pitting sites.
- Chlorinated solvents. Brake cleaner plus arc heat can generate phosgene.
- No interpass control. Overheating hurts sensitization, distortion, and duplex balance at once.
- Wide weaves. They widen the HAZ and buy nothing.
- Skipping tint removal. A straw-to-blue root is a chromium-depleted root.
- Welding over dirty tacks. Tacks are part of the finished weld.
- Treating the WPS as paperwork. If the welder at the pipe cannot read the purge limit from it, it is decorative.
What It Actually Costs
Purge time is the number most shops underestimate. Filling a spool, waiting for the oxygen reading to drop, and holding the dam in place can take as long as the welding itself — and that is the price of a sound root. Orbital welding inverts the math: high setup and equipment cost against consistent travel speed, minimal rework, and fewer highly skilled hours per joint, which pays back fastest on repeatable sanitary and high-purity work. Pickling and passivation deserve their own line item rather than being buried in “finishing.” Rework is the expensive one, because cutting out a root means re-purging, re-welding, re-cleaning, and re-inspecting a joint you already paid for once.
A Prioritized Action List
- Buy a calibrated oxygen analyzer and require a documented reading before every purge.
- Separate and mark stainless tooling — brushes, discs, wheels, files.
- Ban chlorinated solvents from the shop floor outright.
- Write purge, interpass, and cleaning limits into the WPS in numbers a welder can check.
- Add fume extraction at the arc and baseline air monitoring for hexavalent chromium.
- Formalize tint removal, pickling, and passivation as hold points, not afterthoughts.
- Track rework hours by cause for one quarter and let the data set your next priority.
The Bottom Line
The 2-inch 316L spool from earlier in this guide did not fail because the welder lacked skill. It failed because nobody measured the purge, nobody kept carbon steel away from the joint, and nobody removed the tint before the line went into service. A stainless procedure is not a document that satisfies an auditor — it is a list of the specific controls that keep a weld’s inside as sound as its outside. Get the purge right, control the heat, keep the tools clean, and clean up properly afterward. Do those four things and the failures stop being mysterious.
Frequently Asked Questions
Do you always need a written WPS to weld stainless steel pipe?
For code-governed work under ASME B31.3, B31.1, or a pressure vessel code, yes — the fabrication code requires welding to be performed under qualified procedures. Even in non-code shop work, a written WPS is the cheapest way to keep purge, heat input, and interpass limits consistent between welders and shifts.
What oxygen level should I purge to for stainless pipe?
General industrial work often accepts below roughly 2%, but that will still produce visible tint. For good color aim below 100-200 ppm, and for sanitary, pharmaceutical, or high-purity lines target 50 ppm or lower. Your procedure should state the number and the measurement point.
Can I weld stainless steel pipe without back purging?
You can, but the root will oxidize, and on small-diameter or thin-wall pipe you cannot reach it to clean afterward. Unpurged roots lose corrosion resistance and create a crevice that traps product or moisture. Flux-coated or self-shielding root options exist but are not a substitute for purge on critical lines.
Does passivation remove heat tint from a weld?
No. Passivation removes free iron and restores the passive film, but it does not dissolve the thick chromium oxide that forms as heat tint. Tint has to be removed mechanically or by pickling first; passivation comes after.
Why does my stainless weld rust even though it passed visual inspection?
Visual inspection only sees the outside. Rust usually traces back to free iron contamination from carbon steel tooling, an oxidized or sugared root, inadequate cleaning before welding, or arc strikes that were never blended out. Ferroxyl or copper sulfate testing and root-side inspection reveal what the eye misses.
Is preheating required for stainless steel pipe?
Generally no. Austenitic stainless steels are not preheated for metallurgical reasons the way carbon steel is, and preheating can push the joint toward the sensitization range. Gentle warming is sometimes used only to drive off moisture or condensation, and any preheat must be specified in the WPS.
What is a safe interpass temperature for 316L pipe?
Under ASME construction the interpass temperature for austenitic stainless is limited to 350 F (177 C), measured a short distance from the joint. Duplex grades are usually held lower, often around 300 F (150 C). Always follow the number written in your qualified WPS.
Why is duplex stainless pipe welding held to tighter heat input limits?
Duplex relies on a balanced ferrite-austenite microstructure. Too much heat input and slow cooling can form brittle intermetallic phases such as sigma; too little heat input and fast cooling leaves excess ferrite with poor toughness and corrosion resistance. Nitrogen in the shielding and backing gas helps restore austenite.
Can I use a carbon steel wire brush on stainless?
No. A brush that has touched carbon steel embeds free iron into the stainless surface, and that iron becomes a pitting site. Keep dedicated stainless-only brushes, grinding discs, and cutting wheels, and store them separately.
How do I protect welders from hexavalent chromium fume?
Start with source extraction or local exhaust ventilation at the arc, since engineering controls are more effective than PPE. Where exposure can still exceed the 5 microgram per cubic meter PEL, add respiratory protection, monitor exposure, and follow the regulated area and training requirements of OSHA’s chromium(VI) standard.
Nicholas Harper
Editorial TeamCovering welding processes, equipment selection, and shop-floor practice for WeldZen — with every claim checked against the governing code.
Related Studies
Get The Next Benchmark First
New field studies and calibration sheets, once a week. No spam.