Tack Welding Guide — Types, Sizing, Code Requirements & Best Practices | KUSLA

Everything fabricators and engineers need to know about tack welds — from AWS D1.1 qualification clauses to preheating rules most shops overlook, with sizing tables backed by welding handbooks and code references.

What is tack welding?

A tack weld is a short, temporary weld that holds two or more pieces of metal in alignment before the final weld is laid down. Think of it as stitching fabric with pins before sewing: the tacks keep everything registered so the welder can focus on bead quality instead of fighting fit-up drift.

Tack welds are typically 6–25 mm (¼”–1″) long, depending on material thickness and joint type. They use the same welding process as the final weld — MIG, TIG, or stick — but at lower heat input and shorter arc time. Despite being “temporary,” tack welds are subject to the same quality requirements as production welds in most fabrication codes. A cracked or porous tack can propagate straight into the final weld pass if it isn’t caught.

Tack weld vs. spot weld — the real difference

These two terms get confused constantly, even in fabrication shops. They are fundamentally different processes serving different purposes.

The confusion often starts because MIG welders have a “spot weld” timer mode that produces a single, timed pulse. That arc spot weld is a permanent joint method — it is not a tack weld, even though it looks similar. When a drawing calls out a tack weld, it means a temporary alignment weld that will either be consumed by the final pass or ground away. A spot weld callout means a permanent joint.

Three types of tack welds

1. Standard tack weld

The most common type. A small weld bead placed inside the joint groove or along the fillet line. When sized correctly, a standard tack is consumed entirely by the final weld pass and leaves no visible trace. This is the default for well-fitted joints where alignment is good and gaps are minimal.

2. Bridge tack weld

Used when there is a gap between the workpieces — typically due to fit-up tolerance, warping, or thermal distortion from previous operations. Bridge tacks are placed across the gap to pull the pieces together. They are often alternated on each side of the joint to distribute shrinkage forces evenly and avoid creating a bigger gap at the opposite end. Bridge tacking requires higher skill because the weld pool has no backing material underneath.

3. Hot tack weld

Involves extra heat input and additional filler metal, sometimes followed by peening (light hammer blows) to draw parts together and close large gaps quickly. Hot tacks are a last resort for heavy fabrication and repair work where standard and bridge tacks can’t hold. They sacrifice precision for raw holding force — and because of the higher heat input, they carry more risk of distortion and residual stress. Most quality-critical fabrication codes discourage hot tacking unless alternatives have been exhausted.

Tack weld sizing and spacing by material thickness

No single code dictates universal tack weld spacing. AWS D1.1 sets maximum tack weld size for qualification testing (¼” fillet, 2″ length per clause 6.16.1.20) but leaves spacing to the WPS and engineering judgment. In practice, sizing and spacing follow established workshop guidelines documented in welding handbooks and manufacturer training materials.

✅ Verified Spacing values cross-referenced with JASIC welding handbook guidelines and general fabrication practice. These are workshop recommendations, not mandatory code limits — always defer to your project WPS.

AWS D1.1 code requirements for tack welds

AWS D1.1 (Structural Welding Code — Steel) is the most widely referenced code for structural tack welding in North America. It treats tack welding as a qualification category distinct from welder and welding operator certification. The key clauses:

Tack welder qualification (Clause 6.16.1.20)

A tack welder must qualify by welding one test coupon in each position they will tack in production. The test requires a fillet weld of ¼” (6 mm) maximum size, no more than 2″ (50 mm) in length. The coupon undergoes a fillet break test — if it breaks, the tack welder fails and must retest.

Testing and acceptance (Clause 4.23 & 6.24.1)

Tack welds must have a sound weld profile and be free from cracks, overlap, and undercut exceeding 1/32″ (1 mm). This visual acceptance criterion applies to every tack weld on the job — not just the qualification coupon. In practice, this means: no porosity visible to the naked eye, no cracks of any length, and undercut no deeper than a business card’s thickness.

Incorporation or removal

Tack welds that remain in the completed joint must meet the same quality requirements as the production weld. For complete joint penetration (CJP) groove welds, tack welds are typically ground to a feathered edge or removed entirely before the root pass, to avoid trapping slag or creating a fusion defect at the root. For fillet welds, properly made tacks are consumed into the final weld bead — but cracked or porous tacks must be ground out first.

Preheating for tack welds — the rule most shops skip

This is the single most important tack welding practice that competitors’ guides completely omit. When a WPS specifies preheating for the final weld, the preheat requirement applies equally to tack welds. In fact, the preheat temperature for tacking should be 40–50°C higher than the minimum specified for production welding.

Why higher? Because tack welds are short, they cool much faster than a continuous weld bead. Faster cooling means a steeper thermal gradient and a higher likelihood of forming martensite — the hard, brittle microstructure that cracks under hydrogen embrittlement. The extra 40–50°C compensates for the accelerated cooling rate.

✅ Verified Source: Kobelco Welding Education Center, “Tips for Making Sound Tack Welds,” 2003. Kobelco is one of the world’s largest welding consumable manufacturers and this recommendation appears in their published training materials.

The same Kobelco guidance adds a second critical rule: use low-hydrogen electrodes for tacking thick components of mild steel, high-tensile steel, and heat-resistant low-alloy steel. Low-hydrogen consumables (e.g., E7018 for stick, or ER70S-6 wire for MIG with dry shielding gas) dramatically reduce the diffusible hydrogen available to cause delayed cracking.

Why tack welds crack: the metallurgy

Tack welds are uniquely susceptible to cracking for three reasons that compound each other. Understanding the mechanism makes prevention intuitive rather than prescriptive.

Rapid cooling (air quench effect). A tack weld deposits a small volume of molten metal onto a large, cold base plate. The thermal mass of the base plate acts as a massive heat sink, pulling heat out of the tack far faster than it would from a continuous bead. In hardenable steels (carbon equivalence CE > 0.40), this rapid quench produces martensite — a body-centered tetragonal crystal structure that is extremely hard but brittle. Martensite has near-zero tolerance for hydrogen.

Hydrogen pickup. Moisture on the base metal, humidity in the shop air, oily surfaces, and certain electrode coatings all introduce hydrogen into the weld pool. At welding temperatures, hydrogen dissolves readily in molten steel. As the tack cools, the hydrogen becomes trapped in the lattice. In a soft, ductile microstructure (ferrite-pearlite), trapped hydrogen is tolerable. In martensite, it migrates to grain boundaries and stress concentration points, building internal pressure until a crack initiates.

Residual stress concentration. Every tack weld creates localized shrinkage stress as the weld metal contracts during cooling. Because the tack is short and surrounded by cold, rigid base metal, the contraction is restrained — producing triaxial tensile stress at the start and stop craters. These are exactly the locations where hydrogen-assisted cracking initiates.

The result is the “hydrogen cracking triad”: susceptible microstructure (martensite) + diffusible hydrogen + tensile stress. Remove any one leg of the triad and cracking does not occur. That’s why preheating (prevents martensite), low-hydrogen consumables (removes hydrogen), and proper sizing/sequencing (reduces stress) are the three pillars of sound tack welding practice.

Tack welding sequence for distortion control

The order in which you place tack welds matters as much as their size and spacing. Random tacking — starting at one end and working toward the other — accumulates shrinkage in one direction and pulls the joint out of alignment.

Center-out alternating sequence

The most effective general-purpose sequence: place the first tack at the center of the joint, then alternate between each side, working outward. This distributes shrinkage forces symmetrically and prevents the joint from “walking” to one side.

For a 1-meter long butt joint, a practical sequence might be:

Tack 1: 500 mm (center)
Tack 2: 250 mm (quarter point, left)
Tack 3: 750 mm (quarter point, right)
Tack 4: 125 mm (eighth point, left)
Tack 5: 875 mm (eighth point, right)
Tack 6: 50 mm (near start)
Tack 7: 950 mm (near end)

Skip pattern for long seams

On joints longer than 2 meters, a skip pattern works better: tack every third or fourth position first, then fill in the gaps on a second round. This is essentially the same principle as “backstep welding” applied to tacking — it prevents thermal buildup in any one zone.

Equal-sided tacking for box structures

When tacking a box or frame, tack all four sides equally before completing any one side. If you fully tack one seam first, the shrinkage from those tacks will pull the remaining sides out of square. Tack one corner per side in rotation.

Material-specific tack welding tips

Carbon and low-alloy steel

The most forgiving material for tacking — but also the most dangerous if you ignore preheat on thick sections. Below 6 mm thickness with CE < 0.40, tack at ambient temperature with no special precautions. Above 12 mm or CE > 0.45, preheat 40–50°C above the WPS minimum and use low-hydrogen consumables. Always grind start/stop craters to a smooth profile to eliminate stress risers.

Stainless steel

Stainless is sensitive to heat tint (oxidation discoloration), which indicates chromium depletion and reduced corrosion resistance. Tack quickly with minimum heat input. Use back-purge gas (argon) on the root side for austenitic grades (304, 316) to prevent sugaring. Always use stainless filler — never tack stainless with carbon steel wire, which contaminates the weld and creates a galvanic corrosion site. Clean with a dedicated stainless-only wire brush; carbon steel bristle contamination is a real and common defect.

Aluminum

Aluminum’s high thermal conductivity draws heat away from the tack rapidly, making fusion difficult. Use higher amperage than you think you need, and move fast. Always remove the oxide layer immediately before tacking — aluminum oxide reforms within minutes of cleaning, and it melts at 2,072°C versus the base metal’s 660°C, so an oxide layer will prevent fusion without any visible warning. TIG with AC current and ER4043 or ER5356 filler produces the best results. Autogenous tacks (no filler) on aluminum tend to crack because the thin tack bead solidifies under high shrinkage stress with no filler metal to absorb it.

Tack weld symbols on engineering drawings

This is a topic no competitor article covers, yet it’s essential for anyone reading fabrication drawings.

AWS A2.4 (Standard Symbols for Welding, Brazing, and Nondestructive Examination) does not define a unique symbol specifically for tack welds. Instead, tack welds are called out using standard weld symbols — typically a fillet weld or groove weld symbol — with a note in the tail of the welding symbol or a general note on the drawing specifying that the weld is a tack. Common drawing notations include:

“TACK WELD — DO NOT INCLUDE IN FINAL WELD”  or
“TACK WELD ONLY — TO BE CONSUMED BY PRODUCTION WELD”  or
“TW” in the tail of the welding symbol

Some fabrication shops use a shorthand: a small circle at the intersection of the reference line and the arrow, similar to a spot weld symbol, with “TW” noted. This is not formally standardized in A2.4 but is widely understood in North American structural fabrication.

The ISO system (ISO 2553) similarly lacks a dedicated tack weld symbol. Tack welds are communicated via general notes or work instructions referenced on the drawing.

Pipe tack welding: clock positions and tie-ins

Pipe tacking follows a different logic than plate tacking because the joint is circumferential and the weld position changes continuously around the clock.

Standard four-tack setup

For open-root pipe joints, the standard practice is four equally spaced tack welds at the 12, 3, 6, and 9 o’clock positions (or offset to 11, 2, 5, and 8 o’clock depending on welder preference and the planned root pass start point). Each tack should be approximately ¾” (19 mm) long and must achieve full root penetration — a surface tack without penetration will create a defect when the root pass arrives at that point.

Large-diameter pipe sizing

On large-diameter pipe (typically > 12″), tack weld length scales upward. A common rule of thumb is 5–10% of the pipe diameter. A 24″ pipe would use tacks approximately 1.2″–2.4″ long. More tacks may be needed — six or eight equally spaced — to prevent the root opening from closing under gravity and thermal stress.

Tie-in technique

A “tie-in” is the point where the root bead fuses into a tack weld. Good tie-ins require that all slag is ground off the tack weld ends before the root pass arrives. A common SMAW technique: extinguish the arc by pushing the electrode quickly through the keyhole at the tack, then withdraw. This ensures the keyhole size is maintained and the next tie-in will fuse completely. Poor tie-ins are the number one source of root defects in pipe welding.

Common mistakes and how to avoid them