H beam gantry SAW welding defects explained: porosity, slag, undercut and distortion

Why does porosity appear in H beam submerged arc welds even when the parameters look correct?

In SAW, over 70% of porosity cases trace back to flux, not to the electrical settings - and the arc is buried, so you cannot see the defect forming. Three causes dominate. (1) Damp flux: granular flux is hygroscopic and absorbs moisture from shop air. Moisture dissociates in the arc and releases hydrogen straight into the pool. Re-bake agglomerated flux at roughly 300-350 degrees C for 1-2 hours before use, store it in a heated holding oven, and never leave flux sitting in an open hopper overnight or over a weekend. (2) Contaminated joint faces: mill scale, rust, oil, grease and marking paint all outgas under the arc. Wire-brush or grind both flange and web within about 20 mm of the weld line, and strip primer from that band. (3) Flux depth too shallow: if the arc flashes above the flux blanket the pool is exposed to the atmosphere. Aim for a flux layer that fully buries the arc - typically 25-35 mm of loose depth - and increase depth when you raise voltage or current, because a longer or stronger arc blows the blanket aside. On a gantry machine, an inconsistent flux feed is also a cause: check that the hopper valve delivers a steady head ahead of the arc at your travel speed.

How do I distinguish slag inclusions from lack of fusion, and fix each one?

Both hide under the flux and both show up only at NDT, but they have opposite fixes. Slag inclusions sit between passes and come from incomplete slag removal, a bead shape that lets slag run ahead of the pool, or a wandering wire that is not centred in the joint. Fix: chip and brush every pass completely before the next one (never assume the arc will re-melt residual slag), adjust the torch so the wire sits in the root of the joint, and raise voltage slightly or lower current to flatten the bead so slag can float clear. Lack of fusion is a cold joint - the weld metal never bonded to the sidewall or the previous pass. Cause: current too low for the section, travel speed too high so the pool runs ahead of the arc, torch angle off, or a root gap and land that vary along the beam. Fix: increase current and/or reduce travel speed to raise heat input, verify torch perpendicularity, and hold the fit-up tolerance - a varying root gap is the classic reason why the same program welds some sections fine and leaves cold lap on others. Practical diagnostic: if defects cluster on one side of the beam, suspect tracking or torch angle; if they appear at random along the length, suspect flux, wire feed or fit-up variation.

What causes undercut along the weld toe and how is it corrected?

Undercut is a groove melted into the base metal at the toe of the weld that was never filled - it reduces the effective throat and creates a stress riser, so inspectors treat it as a geometric defect, not a cosmetic one. Causes, in order: welding voltage too high (a long arc widens the cone and gouges the sidewall), travel speed too fast for the filler volume being deposited (the pool has no time to flow back and fill the toe), and torch angle that puts more heat on one side. Correction: reduce voltage first - it is the single fastest lever on undercut - then reduce travel speed slightly, then re-check torch alignment perpendicular to the joint. Counter-intuitive but common: an over-deep flux burden can also produce undercut, because the confined arc starts digging rather than spreading. On H beam fillet welds, an additional cause is a mismatched leg length: if the machine is set for a 8 mm leg and you actually need 10 mm, operators tend to raise current to compensate and dig the toe. Set the procedure, then hold the parameters - the whole point of a gantry machine is that it repeats exactly.

How do I control flange rotation, camber and bow after welding an H beam?

Every fillet weld shrinks, and four fillets on a beam shrink unevenly - that is what produces flange rotation (the flange tilting up or down), camber (bowing in the web plane) and bow. Control it in this order. (1) Balance the sequence: weld diagonally opposite fillets, not one flange completely and then the other. On a gantry machine the bidirectional feature matters here - welding both sides in one pass keeps heat input symmetric. (2) Control heat input: multiple smaller passes at moderate current beat one heavy pass, especially on flange thickness above about 20 mm; keep interpass temperature within procedure limits. (3) Pre-set against it: some fabricators pre-set a small reverse angle or pre-camber during assembly, typically a few millimetres over the beam length, so the finished beam lands inside tolerance instead of being straightened later. (4) Straighten after welding rather than fighting the weld: an H beam straightening machine - mechanical or hydraulic - corrects flange angle, web-centred distortion and camber in one pass, and is far cheaper per tonne than weld-side heroics. (5) Fit-up matters: a web that is already bowed, or a flange with a rolled-in camber, guarantees a distorted finished beam. Verify plate straightness before assembly.

Why does the arc wander on long welds and how do I stop arc blow?

Arc blow is magnetic, not mechanical - the arc is deflected by residual magnetism in the workpiece or by an asymmetric magnetic field around the arc. It is much more common on DC than on AC, and it gets worse on long, heavy sections such as a 12 m beam, because the return current path becomes long and unbalanced. Symptoms: the bead snakes to one side, penetration becomes uneven, flux is blown away from the arc, and spatter appears on one side only. Fixes, in order of effectiveness: switch to AC if your power source supports it; shorten and reposition the work return lead so the current does not have to travel the length of the beam (clamp it near the welding head and move it as the head travels, or use a twin return); change the welding direction or sequence to backstep; demagnetise the beam if it has been lifted repeatedly by a magnetic crane or has sat in a strong magnetic field. One practical note: if arc blow appears suddenly on a machine that ran fine for months, check the work return connection for corrosion or a loose clamp before you touch the parameters - a high-resistance return is a very common cause.

Which SAW defects should trigger an NDT repair and which can be left?

This is a code question, not a preference question - AWS D1.1 for structural steel and ISO 5817 / EN 1090 give the acceptance criteria, and the project specification decides which NDT method applies. In practice: cracks are never acceptable, in any length, at any location - cut them out and reweld. Undercut beyond the code limit (commonly 0.8 mm deep, or 1 mm in some grades, with limits on cumulative length) must be repaired or blended out. Slag inclusions and lack of fusion are volumetric planar defects and are typically rejectable wherever they exceed the reference size, because they concentrate stress. Porosity acceptance depends on size and distribution - scattered porosity below the reference level is often acceptable, clustered porosity usually is not. Practical shop policy: weld the first beam of each new section and thickness with full visual plus UT or RT before releasing the batch, then sample. On a gantry SAW line the parameters repeat, so once a procedure is proven, subsequent beams will only deviate if flux, wire, fit-up or tracking changes - inspect when one of those changes.

What causes a ropy, high-crowned bead and how do I flatten the profile?

A bead that looks like a rope, sitting high with a narrow flat top, means the arc was confined - almost always too much flux on top of it, or voltage too low. The pool could not wet out to the sidewalls, so the metal piled up. Reduce the flux depth until you see the arc barely covered, and raise voltage in small steps (a longer arc widens the bead). The target for deep-groove work is a width-to-depth ratio of roughly 1.1:1 to 1.4:1 - narrower than that and you risk centreline hot cracking as impurities are pushed to the centre axis during solidification; much wider and you lose penetration and deposit efficiency. The opposite failure, a bead that is very flat and wide with spatter on both sides, means voltage is too high or travel speed too low. Because on a gantry machine the wire and flux are fixed relative to the joint, an unstable bead profile along one beam usually points to a flux delivery problem - a partially blocked valve, a hopper running low, or flux that has compacted and is not flowing freely.

What safety hazards are specific to gantry submerged arc welding in a beam shop?

SAW looks calm - no visible arc, little spatter - and that is exactly what makes it easy to underestimate. Four hazards deserve written controls. (1) Fluoride fume: much of the fume is trapped under the flux, but what escapes carries fluoride compounds from fluxes containing calcium fluoride. Extract at the arc (a travelling hood on the gantry or local extraction following the head) and monitor exposure rather than judging by what you can see. (2) Electrical: SAW runs at high current with the operator close to the work; the return clamp must be secure, and welding cable, torch and flux hose insulation must be inspected because the helmet hides damaged cable from view. (3) Gantry movement: a rail-mounted gantry travels the full 18 m of rail - fit travel limits, an audible or visual travel alarm, and interlocked access so nobody steps into the rail path; never let anyone stand on the beam while the machine is indexing. (4) Hot work on a beam line: flux and slag stay hot long after the arc stops, the beam itself is hot enough to burn through gloves, and a beam flipped by a crane while weld cables are still attached can pull the head or the power source over. Lock out before any maintenance inside the gantry frame.

Technical Specifications of H Beam Gantry Welding Machine

Product Gallery

Applications

The H Beam Gantry Welding Machine is widely applied in steel structure prefabricated buildings, bridge and infrastructure construction, shipbuilding and marine engineering, heavy machinery manufacturing, power plant and industrial facilities, and offshore wind power structures.

Learn more about the H Beam Gantry Welding Machine:
https://www.steelstructurer.com/pid18376825/H-Beam-Gantry-Welding-Machine.htm

For more information about H Beam Gantry Welding Machine and other steel structure intelligent equipment, contact YOMI CNC Cutting & Welding Machinery. Our team of experts is ready to help you find the best solution for your manufacturing needs.