Pipe Welding Procedures: The Complete Guide to WPS, Positions, Passes & Inspection
Note: Pipeline and piping work carries safety, code, and liability responsibilities. Always follow the governing code (e.g., API…
READ ARTICLE →Back stepping is not a one-size-fits-all trick. The core idea — lay a short bead, then start the next one behind it — travels well across every arc process, but the details change with the wire, the electrode, and the metal you are joining. Here is how the technique adapts process by process and material by material.
GMAW/MIG is the friendliest process for back stepping. The continuous wire feed means restarts are quick, and with a little practice your stop points blend cleanly. Keep your contact tip in good shape and trim the wire before each restart so the arc strikes hot and true.
FCAW/flux-core behaves much like MIG but leaves slag you must chip between steps. Back stepping works well here, especially on thicker plate, provided you clean each bead before the next overlap. Polarity matters — get it right with our guide to DCEN vs DCEP for flux core, and for gasless setups see gasless MIG welding.
SMAW/stick demands the most discipline because every step is a fresh electrode start. Back stepping still pays off on long structural seams, but you fight slag entrapment at each restart. Strike slightly ahead of the crater, then back into it. Our E7018 amperage and technique guide covers the restart habits that keep stick steps clean.
GTAW/TIG gives the finest control and the cleanest restarts, making it ideal for thin, distortion-prone work. Set your machine correctly first — see TIG welding polarity and AC or DC for TIG steel — then taper off current at each stop to avoid a crater.
SAW/submerged arc is where back stepping hits a wall. Lincoln Electric notes back stepping “cannot be used economically in automatic welding,” so mechanized processes lean on other sequencing or tandem techniques instead. For high-deposition automation, see robotic submerged arc welding.
Quick Note: The shorter your step, the more restarts you create. On stick and flux-core, that means more slag and crater risk, so weigh distortion control against cleanup time.
Carbon steel is forgiving. Standard back stepping with 50–75 mm steps handles most long seams, and the material’s moderate thermal expansion means you rarely need to go shorter.
Stainless steel is trickier because it expands more and is prone to sensitization if overheated. Research in the Journal of University of Babylon found that in manual metal arc welding of austenitic stainless steel, angular distortion fell 14.32% with a 60 mm bead length and 41.08% with a 40 mm bead — but actually increased 25% at a 30 mm bead. Too short can backfire.
Aluminum is the headline case. Its high thermal expansion makes sequencing especially valuable. A 2024 study in the IIETA journal found that in GTAW of aluminum, back stepping reduced angular distortion by 11.36% at 75 mm stitch length, 14.39% at 50 mm, and 38.64% at 37 mm. Crucially, the microstructures of welds made with and without back stepping were very similar — so you control distortion without hurting metallurgy. For DC aluminum work, see DC welding aluminium.
ESAB also reminds us that aluminum distortion is often driven by overwelding, and that intermittent welds can cut weld volume by as much as 70%. Back stepping pairs beautifully with that mindset: weld less, sequence smarter.
Note: Whatever the process, consistent step length matters more than the exact number. Random steps create random stress patterns — measure, mark, and repeat. Build the muscle memory through deliberate welding practice, and you will find back stepping becomes second nature across every material you touch.
Back stepping is not a magic trick, and it is not free. Every step adds a start, a stop, and a repositioning move, so the real question is whether the extra effort buys you anything. The answer depends on joint length, material thickness, restraint, and how much a warped part actually costs you. Here is a balanced look at what back stepping delivers — and where it quietly wastes your time.
The clearest win is dimensional control. By breaking a long seam into short, staggered segments, heat input spreads along the joint instead of piling up in one growing heat-sink, so contraction forces cancel rather than accumulate. The result is flatter plates, less bowing, and a part that fits the fixture on the first try. Guidance from TWI lists fabrication techniques like back stepping among the most effective ways to prevent distortion before it starts.
Typical reductions in longitudinal distortion run on the order of 30–50%, though the real figure swings with material, heat input, and joint design. For a shop turning out long beams, tanks, or panels, that margin is the difference between shipping a straight weldment and paying a crew to flame-straighten it.
Less rework is the second payoff. Uncontrolled distortion raises production cost through rectification or replacement of distorted weldments, according to a 2024 IIETA study, and that cost compounds across every stage that touches the part — machining, assembly, painting, inspection. Keep the joint straight and you skip all of it. Back stepping also reduces cumulative residual stress, which sharpens fit-up at final assembly and gives repair work a genuine edge: you can re-weld a damaged section without dragging the surrounding geometry out of position.
Every step costs something. Back stepping multiplies arc starts and stops, meaning more repositioning, more stop-start wear on your contact tip, and more arc time spent tying in rather than laying clean bead. On a long seam that adds up, and each tie-in is a potential defect site if you rush the restart.
Wire and electrode consumption rises too: each restart scrubs a little more filler, and in processes like gasless MIG, every stop risks a burn-back that costs you a fresh contact tip or a re-clip. None of that is fatal, but it is real.
The technique also loses its edge as plate gets thicker and more restrained. On short welds — generally under about 6 inches — distortion is already small enough that the extra steps buy almost nothing, and the same holds for very thick, heavily restrained plate that simply will not move.
There is a subtler trade-off, too: fatigue. A 2026 study in Welding in the World found that minimizing distortion through sequence choice can reduce fatigue endurance — the straightest part is not always the strongest one. Where a weldment sees cyclic loading, distortion goals must be balanced against service life rather than chased in isolation.
Finally, back stepping resists automation. Constant reversing direction fights the logic of robotic and automatic systems, so highly automated lines tend to solve distortion with fixturing instead — check out robotic submerged arc welding for how that plays out at high deposition rates.
Quick Note: Back stepping improves straightness but adds cycle time. If your bottleneck is the welding cell rather than the straightening bay, sequencing may cost more than the distortion it prevents.
Run the numbers before you commit. Back stepping tends to pay off when the weld is long, roughly 6 inches or more, and runs along a dimension-critical or visible edge; when the part must hold tight tolerance for assembly or machining; and when straightening or scrapping a bad part costs more than the added cycle time.
If straightening is cheap, tolerances are loose, or the plate is thick and hard to move, skip it. In those cases careful fit-up and balanced sequencing usually do the job with less fuss — a habit worth building into your welding practice from the very start.
A quick rule of thumb from experienced fabricators: if you can hold the finished part to tolerance without back stepping, do not pay for it. When you cannot, the sequence quickly earns its keep.
Back stepping is simple in theory and surprisingly easy to get wrong in practice. The failures below account for most of the trouble, and each one comes down to a cause you can identify and a fix you can apply on the very next pass. None of these are exotic defects—they are simply sequencing habits.
Cause: On thin material, long segments let the local area overheat before you move to the next segment, which defeats the entire purpose. Heat soaks into a small zone, shrinkage stacks up, and the plate still walks. Aluminum and stainless punish long segments hardest, because their expansion and conductivity carry heat further than mild steel.
Fix: Shorten segments on light gauge—2 to 4 inches is a sensible starting point—and let each one cool before working nearby. The thinner the material, the shorter the step.
Cause: Uneven segment lengths create uneven shrinkage patterns, because each bead pulls a different amount as it cools. You get a wave rather than a straight seam.
Fix: Mark every segment with soapstone or a paint marker before striking an arc. A row of tack marks gives everyone on the job the same reference. Consistent pacing matters far more than chasing a magic number—rhythm beats guesswork.
Cause: Gaps or lack of fusion between segments create discontinuities. Cold laps, porosity, and lack of penetration love to hide at every start and stop. A 2026 study found that weld sequences can introduce several stop/start positions that carry a special risk of weld defect development.
Fix: Overlap each segment by at least 1/4 inch, starting the next bead on top of the previous one rather than beside it. If you need a refresher on spotting a cold lap, this guide to good and bad MIG welds shows the visual difference, while overlap welding defect causes are covered in detail elsewhere. Distortion-control guidance from TWI and Lincoln Electric both stress that sequence and restraint choices interact, so tidy tie-ins alone will not rescue a badly planned pattern.
Cause: Every stop leaves a crater, and an unfilled crater is a crack waiting to happen—especially on thicker, restrained, or higher-carbon joints. On code work, an unfilled crater can be a reportable discontinuity rather than a cosmetic flaw.
Fix: Pause at the end of each segment, add a dab, and back-fill the crater before pulling out. Do not simply snap the arc off and move on.
Cause: Back stepping only works when each new segment ties into the already-cooled start of the previous one. Weld the wrong way and you re-heat the same zone, stacking heat instead of spreading it. A common tell is a heat-tinted band that keeps brightening in one spot.
Fix: Confirm direction before every bead, and be deliberate about which end you begin from. Direction discipline is a core skill worth drilling in welding practice.
Cause: Restraint via clamps increases internal stress and can approach the yield strength of the weld metal; excessive restraint risks solidification cracking. Restraint often does more harm than good when the sequence is poor.
Fix: Use the minimum restraint needed for fit-up, allow movement where the design permits, and let sequencing handle what clamps cannot.
Cause: Habit. Welders often reach for back stepping when the job actually calls for skip or balanced welding. The sequences solve different distortion problems.
Fix: Match the sequence to the joint and the distortion mode. If the goal is spreading heat along a long seam, skip or overlapping techniques may be the better tool.
Cause: Extra passes add heat, not strength. Piling on beads to feel safe feeds distortion and raises the odds of the overlap defect.
Fix: Size the weld to the drawing and keep the bead count to the minimum that meets the design. More metal is not more margin.
Quick Note: Treat every tie-in as a mini weld test. Clean the stop, grind or wire-brush the crater, and inspect with good light and—where the code demands it—magnification or dye penetrant. If the tie-in looks suspect, cut it out now; it will always be cheaper than a repair later.
Back stepping looks like a small trick, but the shops that depend on it treat it as standard practice. Once you see where it shows up, you start noticing it everywhere long seams need to stay straight.
In shipbuilding, distortion is not a cosmetic issue; it is a fit-up nightmare. NSRP shipbuilding guidance states that the back-step process prevents “plate walking” and reduces distortion, and it recommends back-step procedures on thin-plate radius corners of 7/16 in or less. On a panel line, a walking plate throws off every downstream stiffener. These thin panels warp quickly, and back stepping is one of the few sequence tricks that scales from a single patch to a full panel line.
Pressure vessel and pipe fabrication lives or dies by roundness and alignment. A girth seam that ovalizes or a long shell course that bows will not fit the next component, so fabricators sequence welds to balance shrinkage around the circumference and pull toward the intended shape. In tube and fitting work, that balance matters just as much: crews rotate the workpiece and split welds into opposing quadrants so the joint closes evenly, a habit worth pairing with a solid weld tube fittings guide.
Aerospace takes it further. NASA weld modeling found that final pass location and balanced double-V sequences strongly affect residual stress and fatigue resistance in flight hardware — a reminder that sequence is a structural variable, not just a straightness one. If you work in a controlled environment, read up on welding processes used in precision manufacturing.
On trailer and chassis frames, back stepping keeps long rail-to-crossmember beads from pulling the frame into a twist. In thin-gauge sheet metal — enclosures, panels, ducting — a fast back-step sequence is often the only thing standing between a flat product and a pretzel. In field repair of finished parts, where you cannot restraighten or re-machine, a cautious back-step sequence protects the geometry you are trying to save.
Back stepping is powerful alone, but it works best in combination with other shrinkage-control tools.
Lincoln Electric identifies back stepping as one of nine core shrinkage-control methods alongside minimum weld size, balanced welding, presetting, and stress relief. Treat these as a toolbox, not a menu of one — and watch your heat input, because a hot, fast pass still dumps more shrinkage into the part than a controlled one. Beginners can dial in those basics with a solid wire welder guide.
Quick Note: The principle never changes — you are not trying to prevent shrinkage, you are controlling where it happens and in what order. The metal will move; good sequencing just decides the direction.
Remember the banana joint from the start? Nothing was wrong with the arc, the angle, or the travel speed. The sequence did it. Back stepping breaks a long bead into short, opposed segments so each pass pulls against the one before it, keeping the seam straight by letting shrinkage cancel itself out rather than pile up in one direction. That single insight — sequence over brute force — is what separates a weld that looks good on the bench from one that still fits after it cools.
Learn it on scrap, build the rhythm, and it becomes second nature. Pair it with solid fundamentals — welding practice pays off fast here — and set up a workspace that supports careful fixturing by reviewing how to plan and set up welding facilities. Control the shrinkage, respect the sequence, and your long welds will finish as straight as they started.
Back stepping means breaking a long weld into short segments and running each segment in the opposite direction of your overall travel. The joint still fills in one direction, but each bead ties into the cooled bead behind it, which spreads heat input and limits cumulative shrinkage.
No. When the tie-in points overlap properly (about 1/4 inch), a back-stepped weld has the same soundness and strength as a continuous bead. The main trade-off is speed, because of the extra starts, stops, and repositioning.
For most structural steel work, 3 to 5 inches works well. Use 2 to 3 inches on thin material to avoid overheating, and 5 to 8 inches on thick plate that dissipates heat quickly. The goal is to let each segment begin cooling before you weld the next one.
No. Skip welding places short beads in a planned, evenly spaced pattern with gaps that are filled later, and every bead runs the same direction. Back stepping runs each bead opposite to overall travel and produces one continuous weld.
Back stepping (backstep sequence) is a longitudinal weld sequence where beads run opposite to the progress of welding. Step backing is a different technique used mainly when welding grooves, where you deposit a short bead and then step back to a new start point, building up passes in a controlled pattern.
It is usually unnecessary on short welds under about 6 inches, on thick and heavily restrained plate that barely moves, and in automatic or mechanized welding where extra starts and stops are impractical. It also should not be used as a substitute for proper joint design and minimum weld size.
Yes. Because aluminum has high thermal expansion and distorts easily, sequencing matters a lot. Research on GTAW aluminum found that shorter back-step segments cut angular distortion significantly, with a 37 mm stitch length reducing it by about 38% compared to a continuous pass.
No, and the best results come from combining them. Back stepping manages longitudinal shrinkage, while clamps, fixtures, tacks, presetting, and balanced welding manage transverse and angular movement. Using both together reduces distortion far more than either approach alone.