Troubleshooting guide › Weld defects › Lack of fusion

Lack of fusion in welding: causes and how to fix it

Lack of fusion (cold lap) is weld metal that lies on the base metal or on the previous pass without actually melting into it. The bead can look fine from outside, which is why it is dangerous: the joint fails in a bend test or in service. It is almost always a heat problem: not enough energy, or energy pointed at the wrong place.

Seen with: MIG / MAG in short-circuit mode on thick plate is the classic case. Also ARC with too low current, and any multi-pass weld where the previous bead was not cleaned.

Reviewed by the Cruxweld technical team, October 2026. Settings are starting points to confirm on a test piece; acceptance is decided by the code or drawing the job is built to. Checks marked “call” are for our technical team.

What it looks like

From outside, a bead that looks piled up and rounded, with a sharp edge where it meets the plate instead of a smooth blend. In a cut section, a dark line between the weld and the plate, or between passes. A bend test opens up along that line.

bead not melted into plateILLUSTRATION · weld cross-section

Illustration, not a photo. Orange marks the defect.

Causes, most likely first, and the fix for each

1Current or voltage too low for the plate thickness

The arc cannot melt the plate surface, so the filler just sits on it.

Fix: Raise current. For MIG above 6 mm plate, use 1.2 mm wire at 200 A or more, or move to spray transfer with mixed gas.

2Travel speed too slow with a big pool

The pool runs ahead of the arc and the arc heats the pool instead of the plate.

Fix: Speed up so the arc stays at the front edge of the pool.

3Wrong torch angle or aim

Arc pointed at the pool or at one plate only.

Fix: Aim at the leading edge of the pool, at the joint line. On a fillet, split the arc equally between the two plates.

4Mill scale, rust or the slag of the previous pass

The arc has to get through the layer before it can fuse.

Fix: Grind to bright metal; chip and brush every pass before the next one.

5Joint too narrow or bevel too small

The arc cannot reach the root or the side walls.

Fix: Open the bevel to 60 degrees included angle for single-V; keep a root gap of 2-3 mm.

6Weave too wide

Edges of the weave are reached only by the cooler part of the arc.

Fix: Limit weave width to about 3 times the electrode or wire diameter.

What changes by process

ProcessTypical reason on this processWhat to do
MIG / MAGThe most common MIG defect on thick plate. Current too low, arc pointed at the pool instead of the joint, travel too slow so the pool runs ahead of the arc, mill scale not removed.Raise current and voltage together, keep the arc on the leading edge of the pool, speed up, grind the joint.
ARC / MMACurrent too low, electrode too large for the position, slag running ahead of the pool, bad weaving.Current in the upper half of the range, smaller electrode in vertical, keep the slag behind.
TIGNot enough current, too much filler dumped at once, too fast.Make the pool wet both plates before adding filler.
Submerged arcWire off the joint centre, voltage too high for the groove, root gap closed.Align the wire, lower voltage, open the root or use a backing.
Spot weldingA “stick weld” with no nugget: current too low, time too short, worn or mushroomed tips, sheets not clean.Raise the tap one step or add weld cycles, dress the tips to the right diameter.

Starting settings that avoid it

MIG / MAG, 1.2 mm solid wire on 10 mm mild steel, fillet. A starting point only; confirm on a test piece of the same plate.

GasAr/CO2 mix for spray, or CO2 at higher current
Current / voltageabout 240-300 A / 26-30 V
Travelarc at front of pool, 25-35 cm/min
Torch angle10-15 degrees push, aimed at the joint line
Multi-passclean each pass; overlap by one-third

Typical values for mild steel. Your wire, plate condition and machine will move them. Our technical team can set them with your operator on a video call.

How to check for it

Lack of fusion hides under a good-looking bead. Checks that find it:

  • Macro test: cut a cross-section, polish, etch with nitric acid in alcohol, and look for a dark line between bead and plate.
  • Fillet break test: hammer the fillet flat; a fused weld tears through the weld metal, an unfused one peels off the plate.
  • Ultrasonic testing finds it inside a butt weld.
  • On a spot weld: peel test, the nugget should tear a button out of one sheet.

Prevention checklist

  • Match wire size and current to plate thickness; do not run 0.8 mm wire on 10 mm plate.
  • Bevel and gap thick joints so the arc can reach the root.
  • Clean every pass before the next.
  • Do a bend test on a sample joint when you change plate thickness or settings.
  • Keep the arc on the leading edge of the pool.

Questions we get asked

The bead looks perfect. How can it be unfused?

A bead that only sits on cold plate can look smooth and even. Fusion needs the arc to melt the plate, not just the wire. Low current with a high wire speed gives exactly this: a fine-looking bead that peels off in a break test.

Is lack of fusion worse than porosity?

Yes. A pore is a round void; lack of fusion is a flat crack-like gap that concentrates stress. Most codes reject it at any size.

Can preheat help?

On plate over 20 mm, or on cold plate in winter, a light preheat of 50-100 degrees C helps the pool wet the plate. It does not replace correct current.

Does the machine matter?

This is the defect that a small machine causes on big plate. Check the current at 60 % duty cycle on the spec table, not the peak current, when choosing a machine for thick work.

Still stuck?

Send a photo of the weld or cut, your machine model, consumable and plate thickness. Our technical team replies on WhatsApp with the likely causes and the next checks to make.

45+ years of welding expertise · GeM Registered OEM Manufacturer · Supported Worldwide

WhatsApp chat