A procedure qualifies at a certain heat input. In production the welder runs a little faster to keep up, the current drifts up, and the joint passes visual inspection but fails a bend test or cracks in service. Heat input is not a paperwork number. It sets cooling rate, hardness, and toughness, and it changes the moment current, voltage, or travel speed moves.
What heat input means
Heat input is the energy delivered per unit length of weld, expressed in kilojoules per millimeter or per inch. It combines the arc energy, voltage times current, with the travel speed. The formula multiplies voltage by current by a process efficiency factor, divided by travel speed. Higher current or voltage raises heat input; faster travel lowers it for the same settings.
The efficiency factor accounts for energy lost outside the arc: submerged arc is high, processes like MIG and stick somewhat lower. Codes specify the factor for each process, so the same measured settings give slightly different heat inputs by process.
Why it controls cooling and hardness
After the arc passes, the weld cools into the surrounding plate. How fast it cools determines what microstructure forms in the heat-affected zone. Fast cooling, from low heat input on thick, cold plate, can form hard, crack-prone microstructures in hardenable steels. Slow cooling, from high heat input or preheat, gives softer, tougher material but can reduce strength and produce larger grains.
This is the trade-off. Raising heat input to avoid cracking can lower toughness and strength in some steels, while keeping it low for strength risks hardening and hydrogen cracking. The procedure is qualified within a range because both extremes cause problems.
Travel speed changes more than productivity
Speeding up at fixed settings lowers heat input and cools the weld faster. It also reduces weld size, so a welder who travels faster to keep pace may under-fill the joint even if it looks acceptable. Conversely, slowing down raises heat input and can overheat and distort the plate. Travel speed is a controlled variable, not a free choice for the operator.
If production needs higher speed, current and voltage must rise to maintain the weld size and heat input within the qualified range. Faster travel alone changes both geometry and material properties.
Preheat works with heat input
Preheat raises the starting temperature and slows cooling, which helps hydrogen escape and prevents hard zones, especially on thick sections and low-alloy steels. It is part of the same cooling-rate control as heat input; a low-heat-input weld can still be safe with adequate preheat. Interpass temperature keeps this condition on multi-pass welds, so the plate does not cool below the required minimum between passes or rise too high from accumulated heat.
Measure preheat at the specified distance from the joint and across the full thickness, not just at one spot. A surface that reads warm can hide cold material underneath on thick plate.
Multi-pass and accumulated heat
Each pass adds heat, and the interpass temperature reflects the buildup. High heat input combined with no waiting overheats the joint, while rapid succession of small passes can still accumulate. On the other hand, letting a joint cool completely between passes loses the preheat benefit. Control interpass temperature within a range rather than welding as fast or as slowly as convenient.
How it varies in the shop
Machine and display readings differ from true values; calibrate meters and check actual travel speed on mechanized welds. Manual welders vary speed and arc length, which changes voltage and deposition. Position welding, joint fit-up, and drafts across the bay alter cooling. Monitor the real parameters rather than assuming a qualified procedure is reproduced because the machine settings look the same.
Hydrogen and cracking
Hydrogen from moisture and contaminants combines with a hard heat-affected zone and residual stress to cause cold cracking hours after welding. Controlling heat input and preheat to avoid hard microstructures, using low-hydrogen consumables, and cleaning the joint address this together. A crack that appears the next day, not immediately, points to hydrogen and cooling rate rather than a simple lack of fusion.
A worked calculation
Suppose a MIG weld runs at 28 V and 220 A with an efficiency factor of 0.8, traveling at 5 mm/s. Arc energy is 6160 W; effective energy is 4928 W; divided by 5 mm/s gives about 986 J/mm, or roughly 1.0 kJ/mm. If the welder speeds to 7 mm/s without changing settings, heat input falls to about 0.7 kJ/mm, a 30 percent drop that can move the weld outside its qualified range. To hold 1.0 kJ/mm at the higher speed, current or voltage must rise enough to restore the energy. Running these numbers with the actual machine settings shows how easily a small productivity change alters the cooling condition.
Differences by process
Submerged arc delivers high heat input and deep penetration with slow cooling, suited to thick plate, but the high energy can reduce toughness if uncontrolled. TIG runs lower heat input and precise control, often needing careful preheat on thick sections. Stick welding varies with arc length and electrode, and flux-cored processes deposit fast at moderate-to-high input. Each process has its own efficiency and practical range, so a heat input qualified for one is not transferred to another without requalification.
Material-specific limits
High-strength low-alloy and pipeline steels often have both minimum heat input or preheat to avoid cracking and maximum heat input to protect toughness, leaving a real but bounded window. Austenitic stainless steel suffers less hardening but is sensitive to distortion and sensitization, where controlled heat and travel matter. Aluminum conducts heat rapidly, so the challenge is establishing fusion without overheating, and conventional cooling-rate logic differs from steel. Set the range from the material code rather than applying a carbon-steel window.
Thickness and heat sink
The same heat input behaves differently on thin and thick plate. Thick sections draw heat away quickly, acting as a large heat sink that hardens the zone; thin plate heats and distorts more easily. Joint geometry, backing bars, and whether the weld is one-sided or two-sided change how fast heat leaves. A procedure qualified on a test coupon of one thickness may not reproduce on a heavier fabrication even with identical settings, which is why codes tie qualification to thickness ranges.
Monitoring and records
Modern welding machines log current, voltage, and sometimes wire feed and time, which supports checking heat input per joint. On manual welds, observe travel speed against marked lengths and verify the meters rather than trusting set values. Record consumable batch, drying, preheat, and interpass alongside heat input, since these together control hydrogen and cooling. A complete record makes a failure traceable; heat input alone does not show whether preheat or cleanliness was missing.
Common errors
Raising wire feed and current without adjusting travel overheats and distorts; speeding travel to fix a hot weld under-fills and hardens it. Assuming thicker plates need less preheat because the machine is large reverses the actual heat-sink effect. Treating visual acceptance as proof of correct heat input ignores the microstructure underneath. The disciplined approach keeps all three variables inside the qualified window and verifies them on the real joint.
Balancing distortion against cracking
Heat input sits between two failure modes. Keep it low and the weld may harden and crack, especially without preheat; push it high and distortion, residual stress, and toughness loss take over. On a real fabrication the answer is rarely the maximum or minimum the process allows but a controlled middle, supported by joint design and fixturing. Balanced welding sequences, smaller symmetric passes, and proper fit-up reduce the temptation to use extreme heat to force a difficult joint. If a joint needs unusually high heat to avoid cracking, examine fit-up, contamination, and material rather than simply raising the window; if it needs very low heat to control distortion, check the clamping and sequence. Treat heat input as one lever among several rather than the single fix for every welding problem.
It helps to brief welders on why the range exists, not just hand over a procedure sheet. An operator who understands that faster travel changes both weld size and cooling is less likely to push outside it to save time, and is more likely to flag a joint that cannot be welded within range rather than forcing it. A short explanation turns the heat input number from a restriction into a shared quality condition.
When in doubt, keep a short record of the measured settings on the first production joint and compare it to the qualified coupon. That single check at startup catches most drift before an entire joint is welded and is cheaper than repairing a batch.
Bottom line
Heat input is arc energy over travel length and sets the cooling rate that controls hardness, toughness, and cracking. Treat travel speed as a controlled variable, pair heat input with preheat and interpass temperature, and verify actual current, voltage, and speed in production. A weld that passes visual inspection outside its qualified heat input can still fail mechanically; the procedure range exists to protect the material, not to limit the welder.
