Interpass temperature

Interpass temperature is a state of the workpiece, not a setting on a machine, which is what makes it awkward to control and easy to violate. It is measured on the weld and the adjacent parent material immediately before the next run is deposited — not while the run is in progress, and not after the joint has been left to cool.

A maximum and a minimum exist for opposite reasons. The minimum — usually the preheat temperature — keeps the cooling rate slow enough to avoid hard, crack-prone microstructure and to let hydrogen diffuse out. The maximum keeps the joint from being held hot long enough to coarsen grain, soften the heat-affected zone or degrade toughness. Both bounds are material and thickness dependent, both are stated in the procedure, and the standards for measuring them are separate from the standards that set them.

The measurement itself is where most disputes start. A contact thermometer reads one point at one instant, and the point a welder can reach is not always the hottest region of the joint. Cadence matters too: on a fast multi-pass sequence a spot check between two runs can miss the interval when the workpiece was above the limit. Non-contact measurement across the whole run answers both objections, at the cost of needing the surface emissivity to be known before an absolute temperature can be trusted.

On thick-section work the practical failure is cumulative rather than dramatic: each run adds heat, the workpiece never returns to the assumed baseline, and the last runs of a joint are welded well above the band the qualification supports.