Intermittent Welding Explained: Stitch, Chain & Staggered Welds (Complete Guide)

What Is Intermittent Welding? Definitions, Types, and Where You’ll See It

Picture a utility trailer rolling down the highway, or a steel handrail bolted to a set of concrete steps. Look closely at the fillet joints and you’ll notice something odd: short beads spaced with even gaps, not one long continuous weld. It looks unfinished. It isn’t. That pattern is intermittent welding, and it’s one of the most deliberate, cost-saving techniques in metal fabrication.

Intermittent welding makes a weld in segments separated by gaps rather than as a continuous bead. The segments carry the specified load while skipping deposited filler metal in between, which cuts material use, arc time, and heat input. AWS A3.0, the standard welding terminology, defines an intermittent weld as one whose continuity is broken by unwelded spaces between weld segments. Simple as that: weld, gap, weld, gap.

The Core Types

Most of what you’ll see in the field falls into a few named patterns. They describe how the segments line up on a two-sided joint.

• Intermittent fillet weld — the umbrella term. Any fillet made in segments with intentional gaps.
• Chain intermittent weld — segments placed directly opposite each other on both sides of the joint. The welds “chain” down the member and the two sides mirror each other.
• Staggered intermittent weld — segments offset so that one side’s weld faces the other side’s gap. This alternation spreads heat more evenly and helps control distortion.
• Stitch weld — a loose shop term. Sometimes it simply means an intermittent fillet weld; other times it describes a tack-and-fill approach used to hold parts while managing distortion.

Chain and staggered patterns share the same segment length and pitch; only their alignment differs. Related but distinct are plug and slot welds, where filler is deposited through a hole or elongated slot in one plate to fuse it to the member below. They share the “not continuous” idea, but they’re a different joint type entirely. Don’t confuse any of these with an overlapping weld, where metal piles up on the surface instead of fusing at the root.

Pitch, Segment Length, and Clear Span

Two numbers describe every intermittent weld callout:

• Segment length — the length of each individual weld bead.
• Pitch — the center-to-center distance from one segment to the next.

The gap between segments is the clear span, which equals pitch minus segment length. A “3-9” callout, for instance, means 3 in of weld on 9 in centers. Codes keep these honest: AWS D1.1 sets the minimum length of an individual intermittent fillet weld segment at 1-1/2 in (38 mm).

Where will you spot it? Trailer frames and crossmembers, handrails and stair stringers, deck plate supports, machinery guards, and non-pressure structural connections where the design load is well understood. In every one of those cases, choosing intermittent over continuous is a design decision made by the engineer, then written on the drawing. It is not a welder’s shortcut, and a welder who decides to skip half a joint on their own has created a defect, not a strategy. Handled correctly, this technique is legitimate engineering that saves filler, time, and rework.

Codes and Standards: AWS D1.1, AISC 360, and the Rules That Govern Intermittent Welds

Intermittent welding isn’t a free-for-all. The moment a joint carries structural load, the gaps between beads change how stress moves through the connection, so the codes step in. An intermittent fillet concentrates force at each segment’s ends and leaves the unwelded gaps carrying nothing, which is fine in the right application and dangerous in the wrong one. That’s why the AWS D1.1 Structural Welding Code—Steel and AISC 360 both put hard limits on where and how you can use them.

What AWS D1.1 Requires

AWS D1.1 is the master rulebook for structural steel, and it treats intermittent fillet welds with real caution. Here are the provisions that matter most on the shop floor:

• Minimum segment length: each individual weld segment must be at least 1-1/2 in (38 mm). Shorter beads cool too fast, lack fusion, and behave unpredictably under load.
• Tied together: intermittent fillet welds must be placed on both sides of the joint and tied together, with a segment on one side roughly opposite a segment on the other.
• Termination: welds must be terminated at the ends of joints and returned around corners, not left hanging in the middle of a plate.
• Restrictions: D1.1 limits intermittent fillet welds in certain compression members and in joints exposed to corrosion or moisture, where crevices between beads trap water and accelerate rust.

The full text is public in the AWS D1.1 code preview if you want to read the clause language yourself.

AISC 360 Chapter J

AISC’s Specification for Structural Steel Buildings covers welding in Chapter J. Section J2.2b requires a minimum fillet weld length equal to four times the nominal weld size, and never less than 1-1/2 in. Minimum fillet weld size itself comes from Table J2.4, keyed to the thickness of the thicker part joined. Effective length rules then decide how much of each weld actually counts toward capacity — odd-sized or badly placed portions don’t count at all. You can dig through the reasoning in the AISC welding FAQs.

AISC also ties intermittent weld design to load path. In a built-up member, the stitch welds must transfer shear between components, so spacing and coverage matter as much as size. The AISC fillet and PJP weld research digs into how these connections behave and fail.

AWS D14.1 and Cyclic Loading

For cranes and material-handling equipment, AWS D14.1 governs intermittent fillet welds and is stricter because these structures see repeated loading. In cyclically loaded structures, the minimum fillet weld size is often specified as 3/16 in (0.1875 in), per AWS D1.1 Table 7.7 notes. A small bead simply can’t survive millions of stress cycles without cracking.

Why Codes Restrict Intermittent Welds

Fatigue is the big one. Under repeated or dynamic loading, the unwelded gaps and the abrupt ends of each segment act as stress risers where cracks initiate. Corrosion is the second: open gaps between stitches form crevices that hold moisture and rust from the inside. The third is joint integrity, where load reversals or tension really want continuous material. That’s why codes steer intermittent welds toward statically loaded, protected, non-critical connections.

Design convention reinforces this. Many engineering specifications, following Blodgett’s Design of Weldments, prefer intermittent welds for any weld over 10 in (250 mm) long, and typically use 50% coverage unless noted otherwise. That balance keeps stiffness and strength while cutting filler and heat.

Drawings and Symbols

None of these rules work if the drawing is vague. Every intermittent weld must be fully defined on the welding symbol: size, segment length, and pitch, the center-to-center spacing. A fabricator left to guess will weld it wrong, and inspection will reject it. Building the skill to read those callouts accurately is exactly what deliberate welding practice is for.

Why Use Intermittent Welding? Cost, Speed, Distortion, and Filler Savings

If a joint only needs to transfer a given load, welding it continuously is money left on the table. Intermittent welding exists because fabrication economics are driven by inches deposited, and every inch you don’t deposit is an inch you don’t pay for in filler, labor, power, or time.

The math is simple: less length means less of everything

Deposited weld metal volume scales with weld length. Hold weld size constant and halve the length, and you roughly halve filler consumption and arc-on time. That’s why so many drawings call out 50% coverage on non-critical joints — it cuts filler and labor close to in half while still tying the members together.

The range is wider than most people assume. Design references such as Blodgett’s Design of Weldments list coverage ratios from roughly 16% up to 75%, each paired with matching segment lengths and center-to-center spacing. A 16% pattern isn’t laziness — it’s a specified schedule with a purpose. The savings stack up across the whole job:

• Filler metal — electrodes and wire bought by the pound
• Labor — arc-on time and welder travel down the joint
• Consumables — shielding gas, flux, and electricity
• Finishing — less spatter to chip, grind, and blend

Reduced heat input tames distortion

The second big win is thermal. A continuous weld dumps heat into the part in one long pass, so the plate expands, contracts, and pulls. Segmentation lets each bead cool before the next starts, lowering peak heat input and reducing residual stress and distortion. That’s exactly why intermittent welds appear on thin-gauge sheet, long structural members, and heat-sensitive assemblies. If you want the mechanics behind arc control and travel technique, our welding practice fundamentals guide covers the ground.

Fewer inches also shrink cleanup. When you deposit less metal you also generate less spatter and fewer overlap defects to chase, so grinding and rework fall along with weld volume.

The strength tradeoff

Intermittent doesn’t automatically mean weaker — it means reallocating metal. When a design is governed by the throat area needed to carry the load, you can increase weld size and reduce length while hitting the same required capacity. A shorter, heavier leg can match a longer, thinner one for less deposited metal. The catch is sized correctly: this is a design decision, not a shop-floor shortcut.

Where the savings are real, and where they’re an illusion

Genuine savings live in static, non-critical joints, distortion-sensitive assemblies, and high-volume production. The illusion shows up under fatigue, corrosion, and sealing requirements. Every intermittent segment has terminations that concentrate stress and can become crack initiation sites under cyclic loading. Gaps also trap moisture, making intermittent welds a poor choice for corrosion-prone or weather-exposed joints, and they cannot provide a fluid-tight seal at all. Standards reflect that caution — AISC guidance on welding and AWS D1.1 both limit intermittent fillet welds where fatigue or corrosion governs.

Automation changes the calculus

On automated and robotic cells, machine time is the currency. Trimmed schedules keep the torch moving and cut arc-on seconds per part, which compounds fast in high-deposition work. Robotic submerged arc and FCAW cells that already lay metal down quickly benefit most from removing unnecessary inches — see our guide to robotic submerged arc welding for how those schedules are programmed.

A quick worked example

Picture a shop building 200 identical brackets, each with a 12-inch fillet and a 1/4-inch leg. Continuous welding deposits 2,400 inches of weld. Switch to 50% coverage at the same leg and that drops to 1,200 inches — half the filler, roughly half the arc-on time. Across the run you save 1,200 inches of wire, plus the gas, electricity, and labor behind it.

Designing Intermittent Welds: Sizing, Segment Length, Pitch, and Coverage

Once you know intermittent welds save time and money, the real question is how much weld you can safely delete, and where the gaps go. Three numbers answer that. Get them wrong, and you’re building a failure point.

The Three Variables That Define an Intermittent Fillet

Every intermittent fillet weld is fully described by three values:

• Weld size (leg): the nominal leg length, the same measurement you’d use on a continuous bead.
• Segment length: the length of each individual bead.
• Pitch: the center-to-center distance from one segment to the next.

Pitch is always greater than segment length. The space between beads, the unwelded gap, is simply pitch minus segment length.

Coverage Percentage

Coverage is what people mean when they say “50% welding.” The formula is simple:

Coverage % = (segment length ÷ pitch) × 100

A 2 in segment on a 4 in pitch gives 2 ÷ 4 = 0.5, or 50% coverage, the classic trailer-frame pattern. Bump the segment to 3 in on the same pitch and you get 75%. On drawings you’ll often see the three-part notation, such as 1/4 x 2 x 4, read as size x length x pitch.

Converting a Continuous Weld Into an Intermittent One

Here’s the design logic every fabricator should understand. A continuous fillet’s strength comes from its effective throat area per unit length. Switch to intermittent and you remove material, so to carry the same load you must add it back. Since you can’t lengthen the beads without shrinking the gaps, the answer is a bigger leg.

The rule of thumb: as coverage drops, required weld size rises. Preserve the effective throat area per unit length and the math works out. Doubling the leg isn’t free, though. It costs more filler per inch and more heat, so the raw savings shrink as coverage drops.

Minimum Length and Pitch Rules

Codes stop designers from shaving welds down to nothing. Under AISC 360, the minimum length of a strength-designed fillet weld is four times the nominal size (4a), or 38 mm. AWS D1.1 sets the minimum intermittent segment length at 1-1/2 in (38 mm). Pitch must be short enough that the unwelded gap never leaves an excessively long unsupported region that can buckle or open under load. As a practical check, keep the gap short enough that the joint can’t be flexed by hand during fit-up.

Weld Size vs. Base-Metal Thickness

Minimum fillet weld size is tied to the thickness of the thinner part. AISC 360 Table J2.4 and AWS D1.1 Table 7.7 both scale the minimum leg up as the base metal gets thicker, because a small bead on heavy plate cools too fast and can crack. Codes prohibit arbitrarily tiny intermittent welds purely to save filler; the leg must meet the table minimum regardless of coverage. That table is the backstop against a fabricator quietly shrinking beads on the shop floor.

Who Sets the Numbers?

The design engineer, not the fabricator, decides weld size, segment length, and pitch. Those values appear on the welding symbol and the drawing detail, and changing them in the shop is a design change, not a shortcut. Executing those callouts consistently is a core welding practice skill.

Long Welds and the Beta Reduction

There’s a catch for long runs. For end-loaded fillet welds longer than 100 times the leg size, AISC applies a reduction coefficient, beta, to the effective length. Stress isn’t uniform along a long weld: the ends carry more load than the middle, so a 40 in bead doesn’t deliver 40 in of uniform capacity. That’s why long, low-coverage intermittent runs behave differently from short ones.

None of this math helps if the beads are executed poorly. Cold lap and overlap can creep into short segments where it’s easy to rush the start and stop. The geometric design is only as good as the welder laying it down.

Intermittent vs. Continuous Welding: Strength, Fatigue, and When Continuous Wins

Ask two welders whether a stitched joint is as strong as a solid one and you’ll get two stubborn answers. Both can be right. Under steady load, a properly sized intermittent weld carries the same design load as a smaller continuous weld. Under a load that reverses, pulses, or vibrates, the story flips hard — and the gaps you saved time on become the exact places cracks begin.

Static Loads: Where Intermittent Holds Its Own

Static strength is about cross-sectional area, not continuity. If a connection is designed so the required weld throat is developed through a series of segments, the joint resists a steady load just as well as a full-length weld of the same effective size. That’s why a shop can run 50% coverage on a bracket that never flexes and sleep fine at night. The sizing math belongs to welding practice; the point here is that nothing about the intermittent pattern itself weakens a static joint, provided the segments are long enough to develop their share of the load and the terminations are sound.

The moment load becomes cyclic, that logic collapses.

Fatigue: The Hard Limit

Fatigue is the slow accumulation of damage from repeated stress. Weld fatigue strength is governed mainly by joint geometry, stress range, weld toe conditions, and penetration — not by the base metal’s strength. AISC’s fillet and PJP weld research and fatigue studies make this clear: upgrading from mild steel to high-strength steel buys you almost nothing in fatigue life, because the crack doesn’t care how strong the parent metal is. It starts where geometry concentrates stress. Independent fatigue work, such as this review of welded joint fatigue behaviour, reaches the same conclusion.

That is the problem with intermittent welds in fatigue service. Every segment termination is a geometric discontinuity — a stress riser. A continuous fillet has two toes running smoothly; a stitched weld has double that, plus abrupt ends where the weld section changes instantly. Those ends become prime crack initiation sites. A crack can start at a segment end, then run along the toe or through the throat until the joint fails, faster with every cycle.

Discontinuous welds also let water sit. Gaps hold moisture against the steel, driving crevice corrosion that eats section and seeds pitting — another crack starter. Leak-tight joints have the same issue in reverse: you cannot seal a pressure boundary that has gaps in it.

Where Continuous Welding Is Non-Negotiable

• Cyclic, dynamic, or vibrating loads — bridges, machinery, fatigue-critical frames
• Pressure-retaining or leak-tight joints — tanks, vessels, pipe
• Corrosive or wet environments, or any joint where moisture can collect
• Members where the weld itself is a fatigue-critical detail

In these cases, codes generally prohibit or sharply discourage intermittent welds. For cyclically loaded structures, a minimum fillet leg of 3/16 in is often specified to guarantee adequate heat input and sound fusion — small, fast segments cool too quickly and risk cold lap or lack of fusion.

When Intermittent Is the Right Call

Continuous weld is waste where it isn’t needed. Use intermittent on statically loaded, non-sealed, drained connections — stiffeners, clips, brackets, and composite-action connections between deck and beam. A quick decision table:

Whatever you choose, the segments must still be clean, well-fused, and free of defects. Learning to tell good welds from bad is the fastest way to catch a stress riser before it becomes a failure. Get the load case right, and the pattern takes care of itself.

Controlling Distortion and Heat Input With Intermittent Welding

Distortion is the quiet tax on every weld. Weld metal shrinks as it cools, and when the part can’t move, that shrinkage turns into residual stress instead. Both are driven by heat input per unit length — the energy delivered per millimeter of joint, set by voltage, amperage, and travel speed. Intermittent welding doesn’t reduce the instantaneous heat of a single bead; it reduces cumulative heat by letting the plate cool during the gaps.

That distinction matters. On thin sheet, long members, and anything with a free edge, stacked heat becomes angular distortion, bowing, and buckling. Split the same weld into segments and the steel gets somewhere to dump that energy.

How Segment Gaps Break the Heat Cycle

Each segment heats a small zone, then stops. The gap works as a built-in cooling break, so inter-segment temperature never climbs the way it does in one continuous bead. A lower peak temperature means less expansion, less contraction on cooling, and less residual stress locked into the joint. The American Welding Society’s structural code ties minimum weld sizes and lengths to base metal thickness for exactly this reason: undersized and overwelded joints both create problems.

Sequencing: Balanced, Staggered, and Backstep

Layout alone won’t tame a stubborn part; the order you run the segments matters just as much. Balanced welding alternates sides — a bead on one face, then the opposite face — so shrinkage pulls against itself around the neutral axis. Backstep welding starts each bead behind the previous one and welds toward the finished end. Skip welding jumps around the joint, letting each area cool before you return. Chain and staggered patterns distribute those segments along the length so the joint doesn’t pull hard toward one end.

Sequence even changes the residual stress left at the weld toe. In an ASME Journal of Engineering Materials and Technology study, the toe measured 80 MPa of residual stress when the final pass landed on the main plate, but 170 MPa when that last pass sat on the attachment side. Fatigue strength differences between pass-sequence variants showed up mainly at low stress amplitude, where initial residual stress matters most.

Restraint, Cracking, and Heat-Sensitive Metals

Heavy restraint is a double-edged sword. Clamps and fixtures hold dimensions, but they also stop the joint from relieving stress as it cools, which raises cracking risk in thick or highly constrained sections. More passes and higher cumulative heat make it worse. On quench-and-tempered or heat-sensitive steels, every extra pass softens the heat-affected zone, so intermittent welds can limit that damage by keeping total heat down — though you still have to respect preheat and interpass limits. If you’re working with hardened material, review how to weld hardened steel before choosing a sequence.

Shop Tips That Actually Work

Clamp and fixture the part, but don’t over-restrain thin work — tack and let it move. Copper backing soaks heat away from the root and helps lock the gap. Always mark your segment layout before striking an arc; eyeballing pitch halfway through a run leads to uneven coverage and stress concentrated in one area. Finally, don’t stack five segments in the same hot zone. Spread the travel and let the thermometer be your guide.

Process setup feeds directly into heat input, too. Polarity changes how much heat goes into the plate versus the electrode — see this guide to polarity for flux core welding — and higher current with a faster travel speed can keep heat per unit length flat while you lay more metal. Good welding practice starts with reading the puddle and knowing when to pause.

Writing and Reading Intermittent Weld Symbols and Callouts

An intermittent weld lives or dies on the drawing. Get the symbol right and the fabricator builds exactly what the engineer calculated; misread it and you get a joint that’s half as strong as intended or twice as expensive as it should be. The notation follows strict, learnable logic under AWS A2.4, so once you know the pattern you can read any callout on a print.

The Basic Intermittent Fillet Callout

On a typical fillet symbol, the left side of the reference line carries the weld size. For intermittent work, two more numbers follow it in parentheses or separated by a hyphen: the segment length first, then the pitch. Pitch is the center-to-center distance between the start of one segment and the start of the next.

So a symbol reading 1/4 (2-4) means a 1/4 in fillet weld laid in 2 in segments on 4 in centers. That leaves 2 in gaps and gives 50% coverage of the joint. The size comes first, the length second, the pitch third. Swap the last two and the weld becomes something the designer never intended.

Chain vs. Staggered Intermittent Welds

When both sides of a joint get intermittent welds, the drawing has to tell the fabricator whether the segments line up or alternate. This is the chain-versus-staggered distinction.

• Chain intermittent: segments on both sides are directly opposite each other, drawn so they align across the joint.
• Staggered intermittent: segments on one side fall in the gaps of the other, giving a zigzag pattern that spreads heat and load along the joint.

Staggered welds show the symbol on both sides of the reference line with an offset notation (often the word “staggered”) so the two sides alternate. Placement above or below the reference line tells you arrow side versus other side, so there’s no guessing which face gets welded.

Coverage Percentage and the Tail

Some drawings skip the math and simply call out a coverage percentage, like “50% intermittent.” That’s shorthand for the same idea, but it’s less precise because it doesn’t fix where segments start. The tail at the right end of the reference line can carry the process, a WPS number, or a reference note, so a tail reading “WPS-12” points the welder straight to the qualified procedure.

Common Callouts in Plain English

Where Prints Get Misread

The most common field errors are predictable. Someone treats 4-2 as 2-4 and welds double what’s needed. A fabricator welds the arrow side when the symbol sat on the other side of the reference line. Or the welder reads the pitch as the gap instead of the center-to-center spacing. Each mistake changes coverage and heat input, exactly what the design was counting on. Clear shop signage and a properly planned layout keep those prints readable and where they belong, as covered in our guide to planning and setting up a welding shop.

The WPS and Automated Weld Schedules

Size, length, and pitch don’t just live on the drawing. Under AWS D1.1 they can be essential variables in the Welding Procedure Specification, meaning you can’t change them without requalification. Match the drawing to a qualified WPS before striking an arc.

Robots read the same data differently. An automated cell turns the symbol into a programmed weld schedule: segment length becomes a duration, pitch becomes a travel distance, and current follows the procedure. Reliable quick-connect welding leads keep both automated and manual stations running without interruption. On the floor, the symbol is the contract, and reading it correctly is the first step to a joint that meets the print.

Applying Intermittent Welding by Process: SMAW, GMAW, FCAW, SAW, and Robotics

Every arc process can run intermittent welds, but each one fights its own battle at the start and stop of every segment. Since an intermittent joint can have dozens or hundreds of those transitions, your process choice quietly decides how clean and consistent the finished work will be.

SMAW (Stick): Restrike and Slag Control

Stick welding is the classic field choice for intermittent work on trailers, gates, and structural repairs. The catch is that every segment begins with a restrike. If you strike an arc inside the crater or on top of slag, you get a cold start, slag entrapment, and cold lap right where the weld needs to be sound. Chip and brush each bead, restart just ahead of the previous crater, then back-step into it to fuse the tie-in.

Electrode size drives how long a segment you can realistically run. A 1/8-inch rod may only give you a comfortable 2 to 3 inches before the coating overheats; bumping to 5/32 or 3/16 lets you stretch segments and cut the number of restarts. If you’re dialing in a specific rod, our guide to 3/16 E7018 amperage covers the settings that keep restarts consistent.

GMAW (MIG): Transfer Mode and Clean Starts

MIG is the natural fit for intermittent welds on thin material. Short-circuit transfer at lower currents keeps heat input and burn-through under control, which is exactly the goal when you’re stitching sheet metal or thin-wall tubing. On thicker steel, spray transfer gives you a smoother toe and better penetration, but it demands higher current and flat, clean positions.

Clean stops matter. Release the trigger with a slight forward push, pause to let the crater fill, and use run-in or wire-feed controls to avoid a crater crack or fish-eye. Start with the wire slightly ahead of the joint and pull back into the seam rather than stabbing into cold metal. Travel speed and gun angle set the bead profile, and a sound toe is what protects fatigue life. If you’re new to the process, begin with our wire welder basics, and keep the gun running smooth with regular contact tip maintenance.

FCAW and Gasless Flux-Cored: Polarity and Slag

Flux-cored runs hot and deposits fast, making it popular for intermittent structural welds outdoors. Polarity is not optional here: gasless self-shielded wire generally runs DCEN, while gas-shielded wire wants DCEP. Get it backwards and you’ll fight spatter, poor penetration, and a slag system that won’t peel. Our guides to gasless MIG welding and polarity for flux core break down the difference. Between segments, chip the slag fully. Trapped slag at a restart is one of the most common intermittent-weld defects.

SAW and Robotics: Where Intermittent Work Shines

Submerged arc welding is the process built for repetitive intermittent schedules. High deposition rates plus mechanized travel mean a machine can lay a precise segment, stop, index, and start again with identical length and pitch every time. That repeatability is ideal for long stiffeners and web-to-flange welds where hundreds of identical segments march down a beam. Robotic cells take it further, programming segment length, pitch, and tie-in overlap digitally so the schedule matches the drawing exactly. See how this works in our guide to robotic submerged arc welding.

Start/Stop Quality Is Everything

Across every process, the same principles hold:

• No cold starts. Begin on hot metal or back-step into the previous crater so the tie-in fuses.
• Fill every crater. A concave crater is a crack waiting to happen; add a brief pause or a small circular motion.
• Overlap intelligently. A short overlap joins segments; too much creates an overlap defect with a sharp, fatigue-sensitive toe.
• Match settings to segment length. If a rod or wire can’t finish the segment cleanly, shorten it or change consumables.

The takeaway: intermittent welding rewards process knowledge. Whether you’re running stick in the field or programming a robot in the shop, the quality of each tiny segment — and every restart between them — is what makes the joint hold.

Common Mistakes, Defects, and How to Fix Them

Intermittent welding fails in predictable ways. Most problems trace back to a design decision, a misread symbol, or a lazy start and stop. Here is the catalog, and the fix for each.

The Nine Mistakes That Sink Intermittent Welds

1. Using intermittent where continuous is required. Fatigue-loaded members, sealed joints, and pressure service demand continuous welds. Welding a 50% intermittent joint as if continuous erases the cost and distortion savings and can overheat thin members — but going intermittent where continuous is mandatory creates a hidden weak point. Fix: check the drawing and governing code before you skip a single inch.

2. Getting the pitch backwards. Pitch is the center-to-center distance between segments. Weld 4 in segments on 2 in centers and they overlap into a continuous weld built the slow way. Fix: pitch must always exceed segment length. A “4 on, 2 off” note means 4 in segments on 6 in centers.

3. Segments too short. AWS D1.1 does not permit intermittent fillet segments shorter than 1-1/2 in (38 mm), and length must also satisfy 4a. Stubby tacks strung along a joint fool nobody. Fix: size segments to both rules, not just the coverage percentage.

4. Cold starts and unfilled craters. These are the classic crack starters, and they sit right at segment terminations. Poor tie-in at those points is one of the most common crack initiation sites in fillet welds. Fix: start on the previous crater, run on and off the plate where you can, and fill every crater before breaking the arc instead of snapping out.

5. No tie-in at joint ends. Intermittent welds must terminate in a continuous weld, or a full-length weld for a specified distance. Fix: read the end-weld note and add that closing segment.

6. Excessive unsupported gap. Too much gap between segments lets thin plates buckle and lets moisture in, which accelerates corrosion. Fix: tighten the pitch or add a continuous end weld and seal it where the environment demands.

7. Weld size mismatch. Minimum fillet weld size is set by the thickness of the thinner part joined, so undersizing to save money violates code. Oversizing on thin sheet invites burn-through. Fix: match the minimum-size table to the actual part thickness.

8. One-sided distortion. Piling every segment on one side pulls the member out of square. Fix: stagger segments side to side and sequence them to balance heat.

9. Misread symbols. Length, pitch, and side all live in one callout. Fix: slow down and verify before the first arc.

The table below maps the most common symptoms to their causes and fixes.

How to Inspect Intermittent Welds

Start with your eyes: even spacing, consistent length, filled craters, and no cold lap at the toes. To sharpen your eye for surface defects like overlap and undercut, see our guide to overlap welding defects and our piece on telling good MIG welds from bad.

Then measure. A fillet weld gage checks leg size in seconds, and a tape confirms segment length and pitch. If the callout reads 3 in on 9 in centers, the tape should agree. Document what you find before paint goes on.

NDT is reserved for fatigue, seismic, or otherwise critical joints. Dye penetrant and magnetic particle testing find surface-breaking cracks at segment ends; ultrasonic testing finds internal flaws. When a joint is code-governed, judge it against the applicable AWS D1.1 acceptance criteria, and check the AISC welding FAQs for structural minimums.

Practical Shop Tips, Inspection, and Conclusion: Getting Intermittent Welding Right

Everything in this guide boils down to one habit: treat intermittent welding like precision work, not leftover work. It is not about welding less; it is about welding exactly where the load demands it. Here is how to get it right on the floor and prove it afterward.

Shop Tips: Layout, Fixturing, and Heat Control

Consistent segment length and pitch are best achieved by marking the joint before welding and fixturing the assembly. Lay out your segments with a soapstone or marker using the same tape every time, then clamp or tack the parts so nothing shifts as the metal shrinks. For distortion-sensitive joints, run a lower-heat process and dial in settings on scrap first.

• Mark both segment and gap lengths along the joint before you strike an arc.
• Fixture or tack strategically: strong tacks at segment ends, light tacks in gaps.
• Balance heat side to side. When both sides are accessible, specifications commonly prefer staggered stitch welds to balance shrinkage.
• Match process and settings to heat control. Lower amperage, travel speed, and correct polarity all reduce distortion.
• Terminate runs correctly: fill craters and tie beads into the joint rather than leaving thin, underfilled ends.
• Document size, length, and pitch on the traveler so the next shift builds it identically.

Inspection: Verifying Size, Length, and Pitch

A fillet weld gage and a measuring tape are the fastest tools for verifying intermittent weld size, length, and pitch in the field. Work through a short checklist:

• Verify leg size against the callout with a fillet weld gage.
• Check each segment length, then pitch, center to center.
• Confirm full fusion and a uniform leg with no cold lap or undercut.
• Confirm the weld is on the correct side and applied in the correct sequence.
• Confirm the finished weld matches the WPS and the drawing.

Verifying the weld against the WPS and drawing is a core part of fabrication inspection and quality control, and it is where loose shop habits surface first. That WPS is usually governed by a code such as AWS D1.1, so know which edition your shop works to. Keep fume and arc exposure controlled per OSHA 1910.252 while you inspect.

The Decision Rule to Remember

Use intermittent welds for statically loaded, drained, non-sealed joints where the engineer permits. Use continuous welds for fatigue, corrosive, or leak-tight service. Coverage is not a feeling; it is a number on the drawing. When in doubt, ask rather than guess.

Making Intermittent Welding Stick

Treat the technique as engineering, not corner-cutting. The savings are real, but they only exist because the gaps are calculated. Practice laying segments of identical length and pitch by following solid welding practice, and set up a clean, well-lit bench area using proven shop layout guidance. The next decade belongs to automation: robotic cells already optimize intermittent schedules for the best balance of strength, distortion control, and deposition. Learn the rules by hand, and you will be ready to program them.

Frequently Asked Questions

What is the difference between a stitch weld and an intermittent weld?

In everyday shop talk “stitch weld” and “intermittent weld” are often used interchangeably to mean a weld made in short segments with gaps. Strictly, an intermittent weld is the code-defined term for a segmented weld, while “stitch” sometimes refers to a temporary or distortion-control technique. Always confirm what is meant on the drawing.

How long must each intermittent weld segment be?

AWS D1.1 sets the minimum length of an intermittent fillet weld segment at 1-1/2 in (38 mm). AISC 360 also requires a minimum fillet weld length of four times the nominal weld size. Whichever is longer governs, and the design engineer specifies the actual length.

What does a weld callout like 1/4 (2-4) mean?

It means a 1/4 in fillet weld deposited in 2 in long segments spaced on 4 in centers, which is 50% coverage. The first number is the segment length and the second is the pitch, the center-to-center distance between segments. Getting them backwards is a common mistake.

Can intermittent welds carry the same load as continuous welds?

For statically loaded joints, yes — a properly sized intermittent weld can develop the same required strength because the weld size is increased to compensate for the reduced total length. For fatigue, cyclic, corrosive, or leak-tight service, continuous welds are generally required instead.

Why are intermittent welds not allowed on some structures?

The gaps between segments create stress concentrations that can initiate fatigue cracks and provide places for moisture to collect and cause crevice corrosion. They also cannot seal a joint, so codes restrict or prohibit them on cyclically loaded, pressure-retaining, and exposed structural members.

Does intermittent welding really save money?

Yes, when the design permits it. Reducing weld coverage from 100% to 50% roughly halves filler metal, arc-on time, labor, and cleanup on that joint, and it also lowers heat input and distortion. Those savings disappear if the joint actually needs a continuous weld for fatigue or sealing.

What is a staggered intermittent weld?

A staggered intermittent weld offsets the segments on one side of a joint so they face the gaps on the other side, rather than being directly opposite each other. It is used to balance shrinkage and reduce distortion when both sides are accessible. A chain intermittent weld places the segments directly opposite each other.

How do I check an intermittent weld during inspection?

Verify the weld size with a fillet weld gage, then measure each segment length and the pitch with a tape to confirm they match the drawing and WPS. Also check for complete fusion, uniform legs, filled craters, and that the correct side and sequence were welded.

Can robots and automated welding do intermittent welds?

Yes. Automated GMAW, FCAW, and SAW systems, including robotic cells, can hold segment length and pitch very precisely, which makes intermittent schedules attractive for repetitive stiffener, web, and deck welds. The programmed schedule must still match the drawing and qualified WPS.

Authoritative Sources & Further Reading

• https://law.resource.org/pub/us/cfr/ibr/003/aws.d1.1.2000.pdf
• https://pubs.aws.org/Download_PDFS/D1_1_2020SecondP_PV.pdf
• https://www.aisc.org/aisc/solutions-center/engineering-faqs/welding
• https://www.aisc.org/media/rp2pj0bm/fillet-and-pjp-welds-final-report-rev1.pdf
• https://pmc.ncbi.nlm.nih.gov/articles/PMC9786587
• https://preserve.lehigh.edu/system/files/derivatives/coverpage/426516.pdf
• https://ej.aisc.org/index.php/engj/article/download/1150/1149/1149
• https://www.osha.gov/welding-cutting-brazing
• https://www.osha.gov/laws-regs/regulations/standardnumber/1910/1910.252