Welded Connection of Galvanized Steel Pipe: Defects and Repair Practice

Where a Galvanized Pipe Welded Connection Goes Wrong

A welded connection between galvanized steel pipes fails for three reasons that are all preventable: the two pipe ends are not coaxial so the joint sits at an angle, one nozzle is oval because of the forming process or a transport impact, or the deposited weld is full of gas pockets that appear as sand holes on the finished surface. Each of these conditions has its own cause and its own remedy, and all three are best dealt with before the first arc is struck.

When two pipe ends are out of line, the wall thickness presented to the arc varies around the circumference, the weld cross section becomes uneven and the joint carries a permanent bending moment. When a nozzle is oval, the groove gap opens and closes as the welder travels round the pipe, so penetration alternates between excessive and incomplete. Sand holes reduce the effective throat of the weld and, in a pressure service, become leak paths.

Preparing the Pipe Ends Before Fit-Up

Where the joint is not in a straight line, or where the pipe meets a nozzle at an inclined angle, the correct first step is to cut a short section off the pipe head and re-machine the end rather than trying to close the gap by force. Cutting back removes the deformed length and gives the welder a true, square end to work from.

The same rule applies to an oval nozzle. Whether the ovality comes from the mill, from storage or from a transport collision, the end is cut back by a short length and the new end checked for roundness across two diameters at right angles. A pipe that cannot be brought back to round within the permitted tolerance is rejected rather than squeezed with clamps, because forced roundness springs back and reopens the gap after welding.

Zinc coating is then removed from the joint area, typically 25-50 mm back from each edge plus the heat affected zone, by grinding or by a stripping disc. This single operation removes most of the porosity risk: zinc melts at about 419 °C and boils at about 907 °C, so under the arc the coating vaporises well below the temperature of the molten pool. If the vapour cannot escape through the solidifying metal it is trapped as the rounded gas pores that operators call sand holes. Edges are then bevelled as required, and the bevel and the adjacent 20 mm are cleaned of oil, scale and moisture.

Alignment, Root Gap and Tack Welding

Fit-up decides whether the joint can be welded at all. A straight edge laid across the joint shows high-low misalignment, which is corrected with external pipe alignment clamps before tacking rather than after. The root gap is set to the value in the welding procedure, commonly 1.5-2.5 mm for a butt joint in pipe of this class, and the gap is checked at four positions round the circumference so that ovality does not go unnoticed.

Verify coaxiality with a straight edge and correct it with clamps before any tack is made.

Set the root gap evenly at all four quadrants.

Tack at the quarter points with a length sufficient to resist distortion but short enough to be fully fused by the root pass.

Inspect the tacks, and cut out and repeat any tack that shows cracking or porosity.

On a galvanized pipe, tacks are made after the coating has been removed locally and they are treated as part of the root pass: they must be clean, fully fused and free of slag before production welding starts. Tacks left with a thin zinc film under them are a reliable source of root porosity.

Welding Practice That Prevents Porosity

Gas pockets form when vapour or gas is generated faster than it can escape, so the practical controls are heat input, travel speed, arc length and gas flow.

Defect	Cause	Corrective action
Sand holes at the nozzle	zinc vapour from the coating entering the pool	grind the coating back 25-50 mm, keep the arc short, do not linger
Uneven junction thickness	out of line pipe ends or an oval nozzle	cut back the pipe head, re-cut square, correct with alignment clamps
Incomplete penetration	gap closed on one side as the welder travels	recheck the gap at four points, use a steady travel speed
Cracking in the tack	contaminated tack or a forced joint	cut out the tack, remove the coating locally and repeat

For manual metal arc welding, low hydrogen basic electrodes stored and baked to the manufacturer's instruction give the lowest hydrogen level in the deposit, while rutile electrodes are easier to strike and re-strike on thin wall galvanized pipe. For tungsten arc welding, DC electrode negative with a well maintained gas flow and a wire fed inside the shield gives the cleanest root. Whatever the process, the arc is kept short, the weld is completed without unnecessary stops, and good ventilation is provided because the zinc fume is a health hazard as well as a weld quality issue. Runs should not be squeezed into a closed root gap, as that forces the vapour into the pool instead of letting it escape to the surface.

Repair, Re-Coating and Pressure Testing

Rejected metal is removed by grinding or by air carbon arc gouging, whichever the procedure permits, until sound metal is reached. The excavation is examined for remaining porosity before welding restarts, and the repaired zone is welded with the same procedure as the original joint. The wall thickness after grinding must still sit within the permitted negative deviation, so the depth of excavation is measured rather than judged by eye.

Because welding destroys the zinc coating locally, every repaired and welded area is re-protected with a zinc rich repair coating applied over the cleaned steel, and the coating is checked for dry film thickness and continuity. The joint is finally proved by the pressure test required by the pipe product standard, with the test pressure held for the specified time and the weld and heat affected zone examined for weeping. Records of cut-back lengths, repair locations and test results are kept with the pipe history, since a heated and re-coated joint behaves differently from the mill coating around it.

Frequently Asked Questions

Q: Why do sand holes appear at a galvanized pipe welded connection?
Because the zinc coating vaporises under the arc and the vapour is trapped in the solidifying pool. Grinding the coating back 25-50 mm from the joint before welding removes most of the risk.

Q: What should be done when two galvanized pipe ends are not in line?
Cut off a short section of the pipe head and re-machine the end square, then bring the ends coaxial with alignment clamps before tacking.

Q: How is an oval pipe nozzle handled?
Cut back the deformed length, re-cut the end square and check roundness across two perpendicular diameters. If it cannot be brought back within tolerance the pipe is rejected.

Q: Is the zinc coating removed before welding?
Yes. Coating is stripped from the joint area and the heat affected zone, typically 25-50 mm from each edge, and restored afterwards with a zinc rich repair coating.

Q: What root gap is used for a butt joint in galvanized pipe?
The gap follows the qualified welding procedure, commonly 1.5-2.5 mm, and it is measured at four positions around the circumference because ovality changes the gap.

Q: How is a weld with sand holes repaired?
The defective metal is ground out to sound metal, the excavation is checked for residual porosity, the joint is re-welded to the same procedure and the area is re-coated and pressure tested.