SS Pipe Welding: The Complete Stainless Steel Pipe Guide
Table of Contents
The inspector knelt beside a freshly completed 4-inch line, shone a light down the bore, and asked the welder one question: “Did you purge it?” The root was charcoal black, crusty, and already rusting in spots — a perfectly sound-looking weld from the outside that would be cut out before lunch. Most SS pipe welding failures are decided long before the arc starts, and this guide walks through every one of those decision points.
SS Pipe Welding Isn’t Carbon Steel With a Shiny Finish
A fab shop outside Houston once shipped a run of 6-inch 316L spools that looked immaculate from the outside: uniform ripple pattern, no undercut, toes wet in nicely. On site, the fit-up crew found the flanges fighting each other by nearly half an inch over a twelve-foot run. The spools were cut apart, re-prepped, re-welded, and re-inspected. The welder had done nothing a carbon steel job would have punished. On stainless, that cost a week of work.
That gap between “it looks fine” and “it is fine” is the whole story of stainless pipe welding.
So here is the plain answer. SS pipe welding is the joining of stainless steel pipe by arc welding processes — most commonly gas tungsten arc welding (GTAW) for the root pass, followed by GTAW, gas metal arc welding (GMAW), flux-cored arc welding (FCAW), or shielded metal arc welding (SMAW) for fill and cap, all performed under a qualified welding procedure with matched filler metal and, for open-root joints, a protected backside.
Notice what that definition insists on: a procedure, the right filler, and a protected root. Skip one and you’re not welding stainless — you’re melting expensive metal.
Why Stainless Refuses to Behave Like Carbon Steel
Three differences explain almost every stainless problem you’ll ever chase.
Heat stays where you put it. Austenitic stainless has a thermal conductivity of roughly 16 W/m·K against 45–50 W/m·K for carbon steel. Carbon steel wicks heat away and forgives a heavy hand. Stainless keeps the heat at the arc, the puddle stays wet longer, and amperage that felt comfortable on mild steel starts over-penetrating.
It moves more than you expect. Austenitic stainless expands about 17–18 µm/m·K versus roughly 12 µm/m·K for carbon steel. High expansion plus low conductivity gives you concentrated heat and a lot of movement, which is exactly how a straight spool becomes a banana.
The protection is chemistry, not color. Stainless resists corrosion thanks to a thin, chromium-rich passive oxide layer that re-forms instantly in air — that layer is the product you’re paying for. Overheat the metal or starve the root of oxygen shielding and you destroy it, sometimes permanently, sometimes only until it’s pickled and passivated. Either way, the shiny surface tells you nothing.
Quick Note: A stainless weld can pass visual inspection on the outside and still be unusable on the inside. Judge stainless pipe from the bore outward, not the cap inward.
The Cast of Characters
- 304 and 304L — the general-purpose workhorses of the austenitic family. Cheap, weldable, everywhere.
- 316 and 316L — 304 with molybdenum added for chloride pitting resistance. The default for chemical, food, and marine service.
- 321 and 347 — titanium- and niobium-stabilized grades for elevated-temperature service where sensitization is a real risk.
- 2205 duplex — roughly 22% chromium, 3% molybdenum, 0.15% nitrogen; high strength and stress-corrosion resistance, but far less forgiving of heat.
- Ferritic grades such as 409 and 439 — thin-wall, non-pressure work like exhaust and light service.
The Documents That Actually Govern the Work
Most stainless process piping in chemical, food, pharmaceutical, and semiconductor plants falls under ASME B31.3 Process Piping, which sets design, examination, and testing requirements. Welding procedures and welder qualifications live in ASME BPVC Section IX. Open-root passes without backing are addressed in AWS D10.11, and structural stainless work falls under AWS D1.6 — both published through the American Welding Society.
Three Pillars, and the Words You’ll Need
Everything downstream hangs on three ideas: protect the root, control the heat, and keep it clean. Those pillars reappear in every section, from prep to inspection.
The vocabulary is manageable. The root is the first pass, establishing penetration at the bore. The hot pass follows immediately to refine that thin root. Fill passes build thickness, and the cap is the visible outer surface. Heat tint is the rainbow discoloration from surface oxidation; sugaring is the severe gray-black version that ruins corrosion resistance. To purge is to displace oxygen inside the pipe with inert gas. Interpass temperature is how hot the joint is when you start the next pass. Behind it all sits the paperwork: the WPS tells you how to weld, the PQR proves that procedure was tested, and the WPQ records that you personally qualified to it.
What’s Ahead
This guide walks that chain in order — alloy and filler selection, cutting and fit-up, TIG root technique, purging and heat tint control, fill and cap process choices, heat input and sensitization, inspection and code acceptance, safety, and finally the shop economics of getting it wrong. For a torch refresher before we start, our guide to TIG welding technique and standards covers the basics, and our breakdown of filler metal selection and classification explains how to read consumable designations without guessing.
It’s written for the people who actually have to make this work: pipefitters, fabricators, maintenance technicians, engineers specifying the joint, and serious hobbyists who’d rather learn it properly than learn it twice.
Know Your Alloy Before You Strike an Arc: 304, 316L, Duplex, and Filler Matching
Picture a fitter walking to the rack with a length of 316L pipe under one arm and a spool of ER308L in the other. Both are stainless, both are the right diameter, and both will run a beautiful bead. Six months later that line is dotted with rust blooms around the welds, because the filler carried no molybdenum and the service water carried chlorides. Alloy selection is the cheapest decision in SS pipe welding and the most expensive one to get wrong, so let’s sort out what’s actually in the rack before anyone strikes an arc.
The Austenitic Workhorses: 304/304L and 316/316L
Most stainless pipe you’ll meet is austenitic. These grades are effectively non-magnetic when annealed, weld cleanly, and tolerate both shop and field conditions.
304 and 304L are the general-purpose choice: 18–20% chromium builds the passive oxide layer that makes the steel “stainless,” and 8–10.5% nickel holds the austenitic structure together. You’ll find them in water lines, food equipment, architectural work, and plenty of utility piping.
316 and 316L add roughly 2–3% molybdenum, and that single addition is the whole point of the grade. Moly improves resistance to chloride pitting and crevice corrosion, which is why 316 is the default for coastal sites, chemical service, and anything seeing salt or chlorinated water. It is not a magic upgrade for high-temperature service, and it isn’t meaningfully stronger at the torch.
Stabilized, Duplex, and Ferritic Grades
321 (titanium-stabilized) and 347 (niobium-stabilized) tie up carbon so it can’t form chromium carbides during elevated-temperature service. Exhaust systems, steam lines, and heat exchangers that run hot are their territory.
2205 duplex runs about 22% chromium, 3% molybdenum, and 0.15% nitrogen, with a pitting resistance equivalent number (PREN, calculated as %Cr + 3.3 × %Mo + 16 × %N) typically at least 34. It offers roughly double the yield strength of austenitic grades and excellent chloride stress-corrosion resistance. It also welds differently, with ferrite balance and interpass limits that demand their own procedure.
409 and 439 ferritics show up where cost and thermal cycling matter more than corrosion resistance: automotive exhaust and appliance or water-heater flues. They’re magnetic, nickel-free, and considerably less forgiving in the weld pool.
The “L” Suffix and Why Carbon Counts
The L in 304L, 316L, ER308L, and ER316L means 0.03% maximum carbon, versus 0.08% for the standard grade. Hold austenitic steel in the 800–1500°F range and carbon combines with chromium into carbides at the grain boundaries, pulling chromium out of the adjacent metal and leaving it vulnerable to intergranular corrosion. Low-carbon grades keep more chromium in solution, so as-welded heat-affected zones hold up without a post-weld anneal. If a cert says 304 rather than 304L, ask why before that pipe goes into a corrosive line.
Filler Metal Selection: Match the Pipe
| Base metal | Matching filler | Why |
|---|---|---|
| 304 / 304L pipe | ER308L | Standard match, low carbon for as-welded corrosion resistance |
| 316 / 316L pipe | ER316L | Carries the molybdenum needed for chloride service |
| 321 / 347 pipe | ER347 | Niobium-stabilized to match high-temperature service |
| 2205 duplex | ER2209 | Balanced chemistry to preserve ferrite/austenite ratio |
| Stainless to carbon steel | ER309L | Over-alloyed buffer against dilution |
These designations follow the AWS A5.9 classification system, where the “ER” prefix means solid wire or rod for arc welding. The classification logic is worth learning properly — our guide to alloy steel welding electrodes walks through how to read those designators without guessing.
Dissimilar Joints: Dilution and the Case for 309L
When stainless meets carbon steel, the arc melts both sides and the pool becomes a blend. The carbon steel contributes iron and almost no chromium, so the deposit drifts toward a lower-alloy composition — that’s dilution. Join 316L pipe to A106 with ER316L and the root pass can land well below the intended chemistry, with hard, crack-prone zones at the interface. ER309L carries roughly 23–24% chromium and 13% nickel, an over-alloyed cushion that still solidifies as sound austenite after picking up several percent iron. That’s also why dissimilar joints are often buttered first, and why carbon migration across the interface becomes a real concern in high-temperature service.
Verify the Material: Certs and PMI
Never assume the stencil, the paint mark, or the box is correct. Stainless pipe should arrive with a mill test certificate traceable to a heat number, and that paperwork should be read before anything is cut. On mixed-material jobs — or any spool fabricated somewhere else — confirm with positive material identification. Handheld XRF or LIBS units separate 304 from 316 in seconds, non-destructively, and they catch the mixed pallet that paperwork missed.
Note: A magnet and a color check will not identify stainless. Annealed 304 and 316 are both effectively non-magnetic, and a piece of 304 will grab a magnet after cold work or welding. Surface color varies with finish and heat history, not grade. Trust the mill certificate and a PMI gun.
Get the alloy and the filler right, and everything that follows — prep, root, purge, heat control — has something solid to build on.
Cut, Bevel, Clean, Fit: The Prep Work That Decides Your Weld
Most stainless pipe defects are not welding defects. They are prep defects that only become visible once the arc starts. A tight gap, a contaminated bevel, or a fit-up forced with a hammer can turn a perfectly good welder into a repair ticket. On stainless, preparation deserves at least as much attention as the root pass.
Cutting Without Contaminating the Cut
A band saw with a dedicated stainless blade gives the cleanest, coolest cut and is the default in most fabrication shops. An abrasive chop saw works, but only with a wheel reserved for stainless. Plasma cutting is fast on thicker walls, provided the process and consumables are rated for stainless and the cut edge is dressed afterward. Waterjet produces an excellent edge with no heat-affected zone, though it is usually reserved for shop work where the equipment is available.
The bigger risk is contamination. A grinding disc that has just touched carbon steel will embed free iron particles into stainless. Those particles rust, stain the surface, and can initiate pitting. Keep stainless grinding discs, wire brushes, and cutting wheels separate, and store them where they cannot be mixed up.
Quick Note: Dedicate a full set of tools to stainless only: grinding discs, flap wheels, wire brushes, files, and even the saw blade. Label them. A shared grinder is one of the fastest ways to wreck a good stainless weld.
Bevel Geometry: Follow the WPS, Not Habit
A standard pipe bevel is 37.5° per side, producing a 75° included angle. Many procedures also call for a 1/16 in (1.6 mm) root land, and typical open-root gaps run about 1/16–1/8 in (1.6–3.2 mm) depending on wall thickness. Those are starting points, not universal law. The qualified welding procedure specification governs the actual bevel angle, land, and gap, and it is the document that matters when the weld is inspected under ASME B31.3.
Gap and land control penetration more directly than amperage does. Too tight a gap restricts the arc and produces lack of penetration at the root. Too wide a gap invites burn-through, excess root reinforcement, and a puddle that is difficult to control. A heavy land can also block fusion at the root face. Check the fit with a gap gauge, not by eye, and recheck after tacking because shrinkage moves the gap.
Cleaning Is a Metallurgical Step
Wipe oils, grease, cutting fluid, and marker residue from the pipe and bevel with acetone or an approved stainless cleaner. Never use chlorinated solvents. Chloride residue left on austenitic stainless can promote stress corrosion cracking, especially once the metal is heated during welding. That is not a cosmetic concern; it is a service-life concern.
Remove heat tint and oxide from the bevel before welding, and wire brush only with a stainless-only brush. If you also work with aluminum, the same dedicated-chemical discipline applies, as our guide to aluminum cleaner for welding explains. On stainless, cleanliness also means keeping sulfur-based cutting fluids and shop dust off the joint face.
Fit-Up and Alignment
Internal misalignment should be kept within the limit stated by the applicable code or project specification, commonly around 1/16 in (1.6 mm) for many pipe sizes, but always confirm the governing document. Use chain clamps and external alignment spiders to bring the pipe into line. Hammering the pipe to close a gap distorts the bore, work-hardens the metal, and can leave a stress concentration that no weld will fix.
Tack placement matters. Put tacks where they will be consumed by the root pass, and make them small enough to blend in completely. Bridge tacks across the gap can hold alignment, but they must be fully removed before the root reaches them. Removable tacks are safer for critical lines because there is no chance of a cold, unwelded tack hiding in the joint.
Wall Thickness Reference for Common Schedules
Wall thickness drives amperage, gap, and pass count. These Schedule 40 values help you plan before you strike an arc.
| Nominal pipe size | Schedule 40 wall |
|---|---|
| 1/2 in | 2.77 mm |
| 1 in | 3.38 mm |
| 2 in | 3.91 mm |
| 3 in | 5.49 mm |
| 4 in | 6.02 mm |
| 6 in | 7.11 mm |
| 8 in | 8.18 mm |
These dimensions follow ASME B36.10M. Use them to sanity-check your settings, not to replace the WPS.
Pre-Weld Checklist
- Material identified against the mill certificate or PMI result.
- Bevel angle, land, and gap checked against the WPS.
- Bevel and adjacent pipe cleaned with acetone or approved cleaner.
- Stainless-only grinding and brushing tools used.
- Fit-up alignment checked; no hammering or forced distortion.
- Tacks placed, sized, and blended correctly.
- Purge setup and gas flow confirmed before the root pass.
Prep is not the glamorous part of SS pipe welding, but it is where consistency begins. Get the cut, bevel, cleanliness, and fit right, and the root pass has a chance to be boring. Boring is exactly what a good stainless weld should be.
The TIG Root Pass: Technique, Amperage, and the Keyhole
The root pass is where a stainless pipe weld is won or lost. It is the first metal to see the arc, the first to be exposed to oxygen from the inside, and the only pass nobody can inspect properly once the cap goes on. GTAW dominates here for three practical reasons: it gives you independent control of heat and filler, it produces a clean arc with no slag, and it lets a skilled hand hold a consistent penetration profile around the pipe.
Why GTAW Owns the Root
Stainless does not forgive heat the way carbon steel does. It conducts heat slowly, so the puddle stays hot and narrow, and it expands aggressively, so every extra amp adds pull and distortion. GTAW lets you dial current down to a few amps, add filler only when the keyhole tells you to, and keep the arc focused. That control is why open roots on stainless pipe are usually TIG, even when fill and cap are done with a faster process.
Machine Setup and Torch Choices
Set the machine for DC electrode negative. Use high-frequency or lift start so you are not scratching the pipe to strike an arc. A foot pedal is fine on a bench, but pipe work usually rewards a thumb control or a remote amperage control at the torch, because you need to trim heat as you move around the joint. Pulse can help on thin walls, but it is not a fix for a bad fit.
For cups, a #6 to #8 ceramic with a gas lens is the usual range. The gas lens gives better coverage and a more stable shield. Grind 2% lanthanated or ceriated tungsten to a smooth taper, and keep it out of the puddle. A dipped tungsten contaminates the root and ruins arc stability. Run argon at roughly 10–20 CFH. Increase flow in a draft or outdoors, but do not open the valve wide and call it protection.
Amperage: Start With the Rule, Then Trim
The old starting rule is about 1 amp per 0.001 in (0.025 mm) of thickness. Treat that as a rough ceiling, not a target. Stainless roots usually need less, especially overhead and on thin wall. Fit-up, land, gap, travel speed, and purge quality all move the number.
| Wall thickness | 2G / flat | 5G vertical-up | 6G vertical-up |
|---|---|---|---|
| 2.77 mm (0.109 in) | 65–80 A | 60–75 A | 55–70 A |
| 3.91 mm (0.154 in) | 80–95 A | 70–90 A | 65–85 A |
| 6.02 mm (0.237 in) | 90–110 A | 80–100 A | 75–95 A |
| 8.18 mm (0.322 in) | 105–130 A | 95–120 A | 90–115 A |
These are starting points only. The WPS, not a chart, sets the final numbers, and under ASME B31.3 that procedure has to be qualified before production welding.
Reading and Working the Keyhole
Establish the puddle at the tack, let it wet both lands, and watch for a small keyhole to open just ahead of the arc. That keyhole is your penetration gauge. Feed filler to keep it a consistent size, and move so it neither closes nor runs away from you. If it closes, you are too cold or moving too fast. If it grows into a large hole, you are too hot, too slow, or starving it of filler. The goal is a steady, controlled opening that follows the torch around the pipe.
Free-Hand or Walking the Cup?
Free-hand TIG is more adaptable. It suits tight spaces, odd angles, and thin wall where you need to react quickly. Walking the cup gives better mechanical consistency on long, accessible joints, but it is harder to learn and easy to drag the cup out of the groove on a poor fit. Neither technique saves a bad gap. Choose the one you can repeat, and practice it until the keyhole stays stable.
Positions, Travel, and Restarts
On 5G and 6G work, root passes are usually run vertical-up. Gravity helps hold the puddle against the joint, and penetration stays more consistent than it does vertical-down. Keep travel around 3–6 in/min (75–150 mm/min) on the root, with a torch angle of roughly 5–15° push. Most roots use 3/32 in (2.4 mm) filler, stepping to 1/8 in (3.2 mm) on heavier walls; the sizing logic in our all-position welding rod guide applies here too.
Never strike an arc on the pipe surface. Arc strikes are stress risers and corrosion sites. Start on a tack or in the joint, taper off at the end of a pass, and feather the crater before restarting. On a restart, begin slightly ahead of the crater, re-establish the keyhole, then wash back into the stop. Tie-in at tacks should be melted fully, not simply covered.
Note: A proper root inside the pipe has a slight, even crown with a light gold or straw tint. A flat, sunken, gray, or black root is a warning sign, no matter how good the outside looks.
Keep the root clean, keep the keyhole honest, and keep your heat under control. That is the whole job.
Purging and Heat Tint: Why the Inside of the Pipe Matters More Than the Outside
A stainless root can look flawless outside and still be scrap. The inside is where the process fluid lives, and that surface depends on a thin, intact chromium oxide layer. Weld it in air and you burn that layer away. You get heat tint at best and a rough, black, sugar-like crust at worst. Neither is cosmetic, and neither can be fixed by wire brushing the bore.
Back purging is the controlled displacement of oxygen from inside the pipe while the root is welded. It is not optional on stainless, and it is not something to eyeball. It belongs in the procedure.
What Happens When You Don’t Purge
At welding temperature, chromium reacts quickly with oxygen. Instead of rebuilding the thin passive film, it grows a thicker oxide scale. Light cases show straw, gold, blue, or purple tint. Severe cases show gray or black sugaring, a rough crust that destroys corrosion resistance and creates crevices where chlorides and process contamination collect.
The damage is not only chemical. A sugared root is a stress riser and a bacteria trap in sanitary service. In aggressive process lines it can become the starting point for pitting and stress corrosion cracking. That is why many specifications treat the root as a critical inspection surface rather than a hidden one.
| Inside appearance | What it usually means | Practical response |
|---|---|---|
| Light straw or gold | Thin oxide, often acceptable in industrial work | Check the governing specification; clean or passivate if required |
| Blue or purple | Heavier oxide and reduced corrosion resistance | Often rejected or remediated by pickling and passivation |
| Gray, black, or sugared | Oxygen reached the root during welding | Cut it out. Pickling will not restore a sugared root. |
Gases: Argon, Nitrogen, and Hydrogen Blends
100% argon is the default purge gas for stainless pipe. It is inert and works across the common austenitic grades. Argon is roughly 1.4 times denser than air, so it fills a pipe from the bottom up and can collect in trenches, pits, and vessel interiors. That density helps purging, but it also makes argon an asphyxiation hazard in confined spaces. Treat any pit or vessel entry under OSHA’s confined space rules with the respect it deserves.
Nitrogen suits some duplex and high-temperature applications, but it is no universal substitute for argon. Argon/hydrogen blends, commonly 95/5, speed purging and produce a brighter root, but hydrogen is not suitable for every alloy and must never be used where the procedure prohibits it. Duplex and some high-strength grades have restrictions, and hydrogen adds a flammability concern. The WPS has the final say; ESAB’s guidance on purge gas selection is a starting point, not a replacement for your procedure.
Mechanics: Seal, Displace, Verify
Seal both ends before you strike an arc. Tape, water-soluble film, inflatable dams, and commercial purge systems all work when used correctly. The goal is to displace air, not to blow gas through the pipe. On smaller pipe, purge flow typically runs around 5–15 CFH, and the space should be displaced for several minutes before welding. Bigger volumes need more time, not more flow.
Do not guess. Use a calibrated oxygen meter. A widely used target for stainless root protection is below 0.1% oxygen, or 1,000 ppm. Critical and hygienic specifications often push toward 50 ppm or lower. Verify at the actual weld zone, not at the gas inlet, and keep the purge flowing until the root has cooled enough to stop oxidizing. For gas selection and safety, this AllGas Welding Supply guide covers the basics.
Reading Heat Tint and Fixing Light Oxidation
Color charts replace opinion with a reference. AWS D18.2 provides reference discoloration levels for the inside of austenitic stainless tube welds, giving inspectors a visual baseline. Acceptance still depends on the governing specification and service. A tint that passes on a utility water line may fail in a pharmaceutical or food-grade system.
Pickling and passivation, following practices such as ASTM A380 and ASTM A967, can remove light heat tint and restore the passive layer. They are remediation for minor oxidation, not a substitute for purge. No acid turns a sugared root into a sound one. If the bore is rough, black, and crusty, the correct repair is removal.
Warning: Never leave a purge dam, foam plug, or water-soluble film inside a line. A forgotten dam can block flow, damage instruments, contaminate product, or force an unplanned shutdown. Count every dam in and every dam out. Tag it, log it, and have a second person verify. This is a documented field failure, not a theoretical one.
Quick Note: A light straw root does not prove the purge was adequate. The first inch of a weld can oxidize while the meter still looks good if the gas was disturbed or the seal leaked. Watch the meter, the seal, and the root together.
A purge log with gas type, flow, oxygen reading, and welder initials prevents the most expensive failures.
Fill and Cap: Choosing Between GTAW, GMAW, FCAW, and SMAW
Once the root is in and inspected, the joint is only half finished. The fill passes give the weld its strength, and the cap gives it its final profile, its corrosion resistance, and its first impression. This is also where welders tend to get comfortable and stop paying attention. That is a mistake. Every pass adds heat, and every extra pass gives the pipe another chance to pull, oxidize, or crack.
The Hot Pass: Fast, Cool, and Often Skipped
A hot pass should follow the root promptly, while the joint is still warm. Its job is simple: fill the thin root section, help refine the root structure, and burn out light surface oxide before heavier fill passes go in. Keep it fast and relatively low-heat. You are not trying to build a mountain of metal in one pass.
Delay matters. Let the root cool completely and you deal with more shrinkage stress, more surface oxidation, and more cleaning before the next pass. That said, never rush a hot pass over an uninspected root. Check the root first. If it is sunken, sugared, or incomplete, fix it now, because the cap will hide it forever.
Choosing the Process Honestly
Manual GTAW remains the standard for thin walls and critical service. It is slow, but the independent filler feed and precise heat control are hard to beat. Pulsed GMAW gives you higher deposition with good control; the pulsing lowers average heat input while keeping the arc stable, which helps on thin stainless and out-of-position work. FCAW earns its keep on heavier wall and shop productivity, though it brings more fume and more cleanup. SMAW with E308L-16 or E316L-16 electrodes is still common for field repairs and heavier walls because it tolerates wind and surface contamination better than GTAW. The trade-off is slag, spatter, and more post-weld cleaning. If you are weighing stick against the alternatives, this breakdown of why SMAW still wins in many field situations is worth a read.
Short-circuit GMAW is a poor choice for a stainless root. It runs cold, promotes lack of fusion, and produces a rough, inconsistent penetration profile. Some qualified procedures allow it for fill, but spray or pulsed transfer is preferred wherever position permits.
| Process | Typical use | Deposition | Key caution |
|---|---|---|---|
| Manual GTAW | Thin wall, critical, sanitary | Low | Slow; heat builds if you linger |
| Pulsed GMAW | Fill and cap, out-of-position | Medium to high | Requires good parameter control |
| FCAW | Heavy wall, shop work | High | More fume, slag, and cleanup |
| SMAW | Field repair, dirty conditions | Medium | Slag removal; starts and stops |
Technique That Protects the Metal
Use stringer beads and keep weave width to roughly three electrode or wire diameters or less. Wide, lazy weaves concentrate heat at the edges, widen the HAZ, and raise solidification cracking risk. Stagger your stop-start points instead of stacking them in one vertical line. That single habit prevents a lot of localized defects.
Interpass temperature is not optional. Many austenitic stainless procedures cap it around 300–350°F (150–175°C) for corrosion service. Check it with a contact thermometer or temp sticks, not the back of your glove. Clean between every pass with a stainless-only brush or dedicated grinding disc, and inspect each pass while it is still visible. Finding incomplete fusion at the fill stage is cheap. Finding it after the cap means gouging.
Aim for a smooth bead profile with a slight convex crown, good tie-in at the toes, and no undercut along the cap edge. Undercut is more than cosmetic: it thins the section and creates a corrosion trap. Sanitary work often demands smooth, crevice-free caps and little to no discoloration. Industrial piping is more forgiving visually, but the acceptance criteria still come from the code and the owner’s specification, not from what looks “good enough.” Filler classifications and their intended service are defined in standards such as AWS A5.9, so match the wire to the procedure rather than to whatever is on the shelf.
Note: A beautiful cap proves almost nothing about the root. A welder can lay a stack of dimes over lack of penetration, incomplete fusion, or a sugared root. Judge the root from the inside, with a light or a borescope, before you cover it.
Fill and cap work rewards patience more than speed. Control the heat, keep each pass clean, and treat every layer as if it will be inspected, because eventually it will be.
Heat Input, Distortion, and Sensitization: The Metallurgy You Can’t Ignore
Stainless punishes heat in a way carbon steel never does. The same amperage that gives you a comfortable, well-fused bead on mild steel can turn a 316L joint into a sensitized, corrosion-prone liability, or warp a spool so badly it won’t fit between the flanges. None of that shows up under a hood. It shows up six months later as a rust streak, or at fit-up as a gap you can’t close. This is the part of SS pipe welding where the welder has to think like a metallurgist for about thirty seconds per pass.
Heat Input: The Number That Predicts Trouble
Heat input is the energy delivered per unit length of weld, and it’s worth memorizing the formula:
Heat input (kJ/in) = (amps × volts × 60) ÷ (travel speed in in/min × 1000)
Run it on a typical TIG root: 100 amps, 12 volts, 4 inches per minute. That’s 100 × 12 × 60 = 72,000, divided by 4,000, which gives 18 kJ/in. Speed up to 6 in/min with the same settings and you drop to 12 kJ/in. Slow to 3 in/min and you’re at 24 kJ/in. Identical machine settings, wildly different heat going into the metal. That’s why travel speed is a variable in the WPS, not a personality trait.
Voltage on TIG is short and stable, usually 10–14 V depending on arc length. GMAW and FCAW run much higher, so the same amperage puts far more energy per inch into the pipe. Moving from one process to another means recalculating, not assuming.
Sensitization: The 800°F to 1500°F Danger Zone
Here’s what happens inside standard austenitic stainless when it gets hot but not molten. Between roughly 800°F and 1500°F (425–870°C), carbon migrates to the grain boundaries and combines with chromium to form chromium carbides. That chromium is now locked up in the carbide, and the metal immediately around the boundaries is left chromium-poor. Chromium is the whole reason stainless resists corrosion, so those depleted zones become the first place chlorides attack. That’s sensitization, and it’s the mechanism behind intergranular corrosion and the familiar “weld decay” along a seam.
Three things limit it. Low-carbon grades like 304L and 316L cap carbon at 0.03% max, so there isn’t much carbon available to form carbides. Stabilized grades, 321 and 347, use titanium or niobium to tie up carbon before chromium has to. The third control is you: keeping heat input low and interpass temperature down shortens the time the metal spends in that window. Multiple passes stacked hot, one after another with no cooling interval, is the classic way to sensitize a perfectly good pipe.
Duplex: A Different Clock
Duplex 2205 doesn’t sensitize the same way, but it has its own thermal enemy. Hold it roughly between 600°C and 1000°C and sigma phase forms, a brittle intermetallic that wrecks both toughness and corrosion resistance. That’s why duplex procedures usually cap interpass temperature near 300°F (150°C), and why many specifications add hardness limits on top. Duplex welds are also checked for ferrite balance, since austenitic weld metal typically targets about 3–10 ferrite number (FN) to resist solidification cracking while keeping toughness. Ferrite measurement follows the AWS A4.2 practice using a calibrated ferrite scope.
Quick Note: A contact thermometer or temp stick between passes is not paperwork. It’s the cheapest corrosion test you’ll ever run, and on ASME B31.3 work the interpass limit is a procedure requirement, not a suggestion.
Distortion: Fighting Physics, Then Working With It
Distortion and residual stress are cousins, not twins. Residual stress is locked-in internal tension from restrained contraction, and it can exist in a spool that measures perfectly. Distortion is the visible movement that happens when the metal is free to go where that stress pushes it. You manage them differently: stress relief addresses the first, sequencing and restraint address the second.
A familiar shop case: a 40-foot run of 2-inch Schedule 40 stainless, welded from one end in a straight line, no backstep, tacks spaced a couple of feet apart. It pulls into a slow bow and the free end lands well off the flange centerline. Nothing was welded badly. The whole line simply shrank toward the start point, one pass at a time, with nothing to balance it. The fix is rarely exotic: balanced sequences, backstep and skip welding, tighter tack spacing, copper chill blocks to draw heat out, and fewer fuller passes instead of many light ones. Allow for shrinkage in your fit-up and let the pipe move where you can control it. For a deeper walkthrough of technique, our guide to controlling distortion and heat in welding covers the same principles from a different angle, since aluminum fights you with high conductivity instead of low.
Stress Relief: When It Helps and When It Hurts
Post-weld heat treatment is not a default step for stainless. Solution annealing, which requires roughly 1900°F+ and rapid quenching, is a mill or specialist operation and rarely practical on a field spool. Stress relief in the sensitization range will do more harm than good on austenitic steel, and on duplex, holding it in the sigma range can destroy the properties you paid for. Follow the WPS and the owner’s specification. When a code does call for heat treatment, it’s usually for a specific service reason, not as insurance.
Inspection, Testing, and Reading the Weld: What Codes Actually Accept
Ask a welder why a joint was run a certain way and you will eventually hear it: “That’s how we’ve always done it.” In a courtroom, an audit, or a client’s root-cause meeting, that sentence buys nothing. A stainless pipe weld is accepted because it was made to a qualified procedure, by a qualified welder, and then examined to a defined standard. Everything else is opinion — and opinions do not hold pressure.
The Qualification Chain: WPS, PQR, and WPQ
Under ASME B31.3, welding must follow a code-accepted approach rooted in ASME BPVC Section IX, which governs three linked documents. The WPS (welding procedure specification) is the written recipe: alloys, filler, joint design, preheat, interpass limits, amperage ranges, purge, and pass sequence. The PQR (procedure qualification record) is the evidence that recipe was tested — a coupon welded, then mechanically or radiographically examined. The WPQ (welder performance qualification) proves the individual can execute it. Three documents, and all three must line up. A welder qualified on 6G 2-inch Sch 40 does not automatically cover 8-inch heavy wall, and a WPS written for ER316L does not authorize ER308L because the box was closer.
Examination: What Gets Looked At, and How Much
Visual examination comes first and covers 100% of production welds. After that, B31.3 sets minimum percentages for radiography and other nondestructive examination based on service category — normal fluid service, severe cyclic, or high-pressure — and owners, engineers, and end users routinely specify more than the floor. Treat the code number as a minimum, not a ceiling. Radiographic examination of pipe welds follows ASME BPVC Section V, Article 2, while penetrant testing falls under Article 6. Ultrasonic testing is increasingly common on thicker walls, and hardness or ferrite checks appear wherever duplex or corrosion-critical service is involved. On top of that, most projects run positive material identification on the components themselves.
Hydrostatic and Pneumatic Testing
The final integrity check on a line is usually hydrostatic: water at 1.5 times design pressure, held and inspected for leaks and permanent deformation, unless B31.3 permits a specific alternative. Water is the safe choice — it stores almost no energy. Pneumatic testing is substituted when water would damage the system, when the line cannot be dried, or when weight and support loads make hydro impractical. It is also genuinely more dangerous: a compressed-gas failure releases stored energy like a bomb, so exclusion zones, remote gauges, and a written test procedure are not optional.
Reading the Weld: A Defect Diagnostic Table
Most stainless rejections trace to a small set of causes, and the defect usually tells you which one.
| Defect | Usual causes |
|---|---|
| Lack of penetration | Tight root gap, low amperage, fast travel, wrong land thickness |
| Incomplete fusion | Low heat, poor cleaning, bad torch angle, filler fed outside the puddle |
| Suck-back / concave root | Too much heat, over-aggressive purge pressure, no filler support |
| Burn-through | Excess amperage, wide gap, slow travel |
| Sugaring / heavy tint | Purge failure or loss of shielding |
| Solidification cracking | High sulfur or phosphorus, excessive heat input, low ferrite in the deposit |
| Chloride pitting after service | Contamination, wrong alloy choice, unresolved heat tint |
PMI and Ferrite: Catching the Wrong Metal
Positive material identification with X-ray fluorescence is the standard field check for mixed material, and it catches the classic error of a 304 fitting installed where 316 was specified. It takes seconds per component and belongs on every job where two grades share a rack. Ferrite content in austenitic weld metal is measured with a ferrite scope calibrated to AWS A4.2, since a little delta ferrite helps resist solidification cracking while too much hurts toughness and corrosion resistance. Specialty alloys raise the stakes further, which is why shops running Alloy 20 welding procedures treat material verification as a separate hold point. In industries like aerospace TIG welding, the same principles apply with tighter acceptance limits and fuller documentation.
Repairs, Re-Inspection, and Repair Limits
A repair is not a re-flow with extra amps. The defect is excavated — mechanically or by grinding — until sound metal is reached, the excavation is examined, and the area is re-welded using a qualified technique under the same WPS controls, purge included. The repair is then re-examined by the same method that found the original flaw. Most specifications cap the number of repair cycles per joint because repeated heating degrades corrosion resistance and accumulates residual stress; after that, the spool is cut out.
Quick Note: Photograph the root before you cap it. A phone shot down the bore, with the weld identification number visible, costs ten seconds and settles most arguments later. Once the cap goes on, that evidence is gone forever.
Safety on Stainless: Hex Chrome, Argon, and the Hazards That Don’t Glow
Stainless steel is often sold as the clean option: bright, corrosion-resistant, and low-maintenance. At the arc, it is not automatically the clean option for the person welding it. The two biggest hazards on a stainless job are usually invisible, and neither one announces itself with a smell or a warning light.
Hexavalent Chromium: The Invisible Cost of Stainless
Welding stainless produces fume containing hexavalent chromium, Cr(VI), a known human carcinogen. Under OSHA’s chromium standard, 29 CFR 1910.1026, the permissible exposure limit is 5 µg/m³ as an eight-hour time-weighted average, with an action level of 2.5 µg/m³. Cross that action level and the standard’s exposure monitoring and medical surveillance obligations come into play. NIOSH recommends a far more protective limit of 0.2 µg/m³ over an eight-hour shift because it treats Cr(VI) as a carcinogen without a safe threshold.
| Hazard | Key limit or threshold |
|---|---|
| Cr(VI) fume, OSHA | PEL 5 µg/m³, 8-hour TWA; action level 2.5 µg/m³ |
| Cr(VI) fume, NIOSH recommendation | 0.2 µg/m³, 8-hour TWA |
| Oxygen deficiency | Below 19.5% oxygen at sea level |
Process choice changes the dose. SMAW and FCAW generally generate more fume and higher Cr(VI) levels than GTAW or GMAW, largely because of flux coatings and higher deposition rates. Stick welding still has a legitimate place in field work, as this guide to the advantages of stick welding explains, but it is not the low-fume choice. GTAW is not a free pass either. Even occasional stainless welding falls under the standard, and grinding or cutting stainless can add Cr(VI) to the air as well. Visible fume is a poor gauge because much of the fine fraction is easy to miss.
Local exhaust ventilation positioned at the arc is the primary engineering control. General room ventilation only dilutes fume after it has already entered the breathing zone. A respirator or PAPR helmet is a secondary line of defense, not a substitute for capture at the source. Rotating workers to stay under an exposure limit is not an acceptable compliance strategy.
Quick Note: An extraction hood two feet from the arc is mostly decoration. Put the capture point close enough to grab fume as it leaves the puddle, and verify airflow before the first bead.
Argon: The Quiet Asphyxiant
Argon is colorless, odorless, and heavier than air, so it collects in pits, trenches, and vessel interiors. OSHA defines an oxygen-deficient atmosphere as below 19.5% oxygen at sea level, which is exactly what a leaking argon hose or an open purge line can create in a low spot. Use a calibrated oxygen meter, ventilate the space, and treat entry as permit-required confined space work under 29 CFR 1910.146. An open shop door is not a monitor, and a hiss is not a measurement.
Heat, Light, and What You Put on the Metal
Stainless’s low thermal conductivity makes it a safety issue as well as a welding one. A stainless pipe holds heat longer than carbon steel, so a piece that looks cool can still cook a glove. Mark hot work, keep combustibles clear, and wear gloves rated for the job. UV from the arc causes arc flash and skin burns, so use proper screens and protection for everyone nearby. Electrical basics still apply: dry gloves, an inspected torch body, and a solid ground connection.
Keep chlorinated degreasers and brake cleaners away from the arc. Heat and UV can decompose them into phosgene and other toxic gases. Use acetone or a dedicated stainless cleaner, and let surfaces dry before striking an arc. The OSHA welding and cutting pages are a useful starting point for the broader hot-work rules.
Safety Checklist Before the Arc Starts
- Position local exhaust at the arc and confirm it is pulling air.
- Fit-test the respirator or PAPR and inspect it before use.
- Test oxygen levels before entering any pit, trench, or vessel.
- Keep chlorinated solvents out of the work area entirely.
- Screen the arc, mark hot parts, and check the ground connection.
- Know the alloy and the procedure before the first tack.
None of this is exotic equipment or paperwork for its own sake. It is the difference between a long career and a short one, and it deserves the same attention you give the root pass.
Common Mistakes, Shop Economics, and the Practical Playbook
Every section before this one comes down to a handful of decisions made at the bench, at the machine, and at the purge line. Here they are in the order they usually go wrong.
The Ten Mistakes That Cost the Most
- Welding without a WPS. Fix: get the qualified procedure in hand before the first tack. A procedure is a legal document, and ASME B31.3 expects you to follow it.
- Guessing the alloy. Fix: read the mill cert and stencil the heat number on the pipe.
- Skipping purge verification. Fix: meter the oxygen at the exit, every time. Guessing by color is not verification.
- Leaving a purge dam in the line. Fix: count dams in, count dams out, and write it on the traveler.
- Using carbon steel grinding wheels. Fix: dedicated stainless discs, stored separately, never shared.
- Running too much heat “to get penetration.” Fix: fix the gap and land instead of the amperage dial.
- Weaving a wide cap. Fix: stringers, and keep any weave to about three wire diameters.
- Not cleaning between passes. Fix: brush and inspect each pass before the next one goes on.
- Capping over a bad root. Fix: stop, look down the bore, and cut it out early when it is still cheap.
- Skipping PMI on a mixed-material job. Fix: XRF the spool before welding, not after the rejection.
The Real Cost of a Cut-Out
A failed weld on a 6-inch stainless spool is never just a weld. Rework means cutting the joint out, re-beveling, re-cleaning, re-purging, re-welding, re-inspecting, and often a slip in the schedule while the crew waits on the NDT bay. On a long run, the spool itself may be scrapped because there is no length left to re-fit. A purge oxygen monitor costs a small fraction of a single cut-out and re-weld on a large-diameter stainless spool, and it prevents the failure instead of documenting it. That is the whole economic argument for stainless work in one sentence: spend the money on prevention, because the repair invoice is always bigger.
The Playbook for a First Stainless Pipe Job
- Confirm material, schedule, and mill certs against the drawing.
- Read the WPS line by line: amps, filler, gas flow, purge level, interpass limit.
- Mock up on scrap of the same schedule, in the same position.
- Dial in amperage and purge on the scrap until the root is even and pale.
- Weld a test coupon with the exact settings you intend to run.
- Cut, bend, or section the coupon and inspect the root and fusion line.
- Only then move to production, and log what worked.
Quick Note: Photograph the root of your test coupon before capping it. A photo library of good and bad roots trains your eye faster than any chart.
If you are still assembling equipment, our buying guide for TIG machines and shop equipment covers what actually matters for pipe work.
Practice Progression and Your Settings Log
Skills build in a known order: flat plate beads first, then 2G pipe, then 5G, then 6G. Six-G is the most demanding fixed-position test because you weld every position without turning the pipe, and it exposes every weakness in torch angle, filler feed, and heat control. Practice deliberately: one session on keyhole consistency, one on restart tie-ins, one on purge discipline. Then keep a log for each joint recording amperage, voltage, travel speed, gas flow, and purge oxygen level. Six months later that notebook is worth more than any generic settings chart, because it reflects your machine, your torch, and your hands.
DIY, Hiring Out, and When to Cut Instead of Re-Flow
Do the mock-ups and the small, non-critical work yourself. Bring in a qualified welder the moment the joint is code-regulated, high-pressure, sanitary, or bonded by insurance. And when a root fails, cut it out. Re-flowing over a defective root rarely helps, because oxidation and lack of fusion do not disappear with more heat, they just get buried under a good-looking cap.
The Three Pillars, One Last Time
Protect the root, control the heat, keep it clean. Everything else is detail.
- Material verified, filler matched, certs on file.
- Stainless-only tools, clean bevel, WPS gap and land.
- Purge metered below 0.1% oxygen before the arc starts.
- Amps trimmed to wall thickness, interpass temperature checked with a gauge.
- Every pass cleaned and inspected, root photographed before capping.
- Settings logged so the next spool goes faster than this one.
Do those six things and the inspector’s question stops being a threat and becomes a formality.
Frequently Asked Questions
Can you MIG weld stainless steel pipe instead of TIG?
Yes, but with caveats. GMAW works well for fill and cap passes and can be very productive with pulsed transfer, but it is harder to control on an open root without backing. Most critical stainless roots are still TIG welded, then filled with GMAW, FCAW, or SMAW depending on wall thickness and the qualified procedure.
Do you always have to purge stainless pipe when welding?
For an open-root, full-penetration weld on any line where corrosion resistance matters, yes. Without purge gas, the hot root oxidizes and forms heat tint or black sugaring that destroys the passive layer and creates a crevice for corrosion. Closed-root joints with backing, or non-critical cosmetic work, may have different rules set by the procedure.
What gas should I use to purge stainless pipe?
100% argon is the default for most austenitic stainless work. Nitrogen is used for some duplex and high-temperature applications, and argon/hydrogen blends such as 95/5 purge faster and produce brighter roots but are not approved for every alloy or service. Always follow the WPS rather than choosing a gas by habit.
What is the difference between 304 and 316L stainless pipe?
316L contains roughly 2–3% molybdenum, which gives it better resistance to chloride pitting and crevice corrosion than 304. That makes 316L the common choice for coastal, chemical, and food-processing service, while 304/304L remains a cost-effective general-purpose option in less aggressive environments.
Why did the inside of my stainless weld turn black?
Black or gray, crusty root surfaces mean oxygen was still present while the root was hot, usually because the purge was too short, the seals leaked, the flow was too low, or the heat input was excessive. That condition is called sugaring, and it typically has to be ground out or cut out rather than polished away.
Can stainless steel pipe be welded to carbon steel pipe?
Yes, and ER309L is the usual filler because its higher chromium and nickel content compensates for dilution from the carbon steel side. These dissimilar joints need tighter heat control and often a specified maximum dilution or hardness, so they should always be made under a qualified procedure.
What amperage should I use for a 2-inch schedule 40 stainless root?
A reasonable starting point is roughly 80–100 amps for a 3.91 mm wall, adjusted for position, gap, and travel speed. That is only a starting point: the qualified WPS and a test coupon on the same material and schedule are what actually set the number.
Do stainless pipe welds need pickling and passivation?
Where full corrosion resistance or hygienic service matters, yes. Pickling removes heat tint and the chromium-depleted layer beneath it, and passivation restores the passive oxide film, following practices such as ASTM A380 and ASTM A967. Pickling is remediation, though — it is far cheaper to prevent heavy tint with proper purging in the first place.
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.