Welding Heat Input Calculator
Arc energy and net heat input against the limit for your material, plus deposition rate — and why slowing down reduces penetration rather than adding it.
The weld
Wire and job (optional)
The three knobs move it equally
A ten per cent change in voltage, current or travel speed moves heat input by about ten per cent each. Travel speed is not a stronger lever — it is an equal one that most people leave uncontrolled while they fiddle with the other two.
Slowing down does not increase penetration
It is the first thing most people try when a weld looks shallow, and past a fairly early optimum it achieves the opposite. Penetration is driven by current — by wire feed speed in MIG. Travel speed governs how much heat goes in per unit length. Crawl along and the puddle rolls out ahead of the arc, so the arc is burning into metal it has already melted rather than digging into cold base material. What you get is a wide, tall, rolled-over bead sitting on the surface, a large heat-affected zone, and less fusion than you had going faster. If a weld is not penetrating, more amps or a narrower gap is the fix.
And "heat input" is ambiguous unless you say which one. Arc energy is V × A × 60 ÷ (travel × 1000). Net heat input multiplies that by a process efficiency, and those run from 0.6 for TIG to 1.0 for submerged arc — a 40% spread. The same weld can be documented at 1.6 kJ/mm or 1.0 kJ/mm depending on a convention nobody states, so a procedure limit of "2.0 kJ/mm maximum" permits materially different welds. Both figures are shown above for that reason.
- Duplex stainless has a floor as well as a ceiling. Nearly every material has a maximum heat input; duplex also has a minimum, because cooling too fast leaves too much ferrite and not enough austenite. It is the one alloy where welding too cold is a genuine defect — and the one case where slowing down really is the right answer.
- Deposition rate and deposition efficiency are different things. Flux-cored wire lays down metal fast and loses a seventh of it to slag. Stick loses nearly two fifths to coating and stubs. Comparing consumables by the kilogram flatters stick enormously, because barely three fifths of what you buy reaches the joint.
- Heat input is only a proxy for cooling rate, and cooling rate is what actually decides heat-affected zone properties. Plate thickness, joint geometry, preheat and interpass temperature all move it without changing the kJ/mm at all, so two welds recorded identically can behave quite differently.
- Wire size has a current range for a reason. Too much current for the wire gives an unstable arc and heavy spatter; too little gives a cold, ropey bead sitting on the surface. Changing wire size is often the fix people reach for last.
How to use
- Pick the process and the material being welded.
- Enter arc voltage, current and travel speed.
- Add wire diameter and feed speed for deposition and consumable figures.
- Check the net heat input against the limit, and fix it with travel speed rather than current.
Frequently asked questions
Does slowing down give me more penetration?
No, and past a fairly early optimum it does the opposite. Penetration is driven by current — by wire feed speed in MIG — while travel speed governs how much heat goes in per unit length. Crawl along and the weld puddle rolls out ahead of the arc, so the arc is burning into metal it has already melted rather than digging into cold base material. The result is a wide, tall, rolled-over bead sitting on the surface with a large heat-affected zone and less fusion than you had going faster.
So what should I change if a weld is not penetrating?
More current, or a narrower gap, or both. If it is burning through or undercutting instead, less current. Travel speed is the wrong knob for either problem, and it is the one people reach for first because it is the only adjustment that does not involve stopping and turning a dial. Joint preparation matters more than most people credit too: a tight square butt in thick plate will never penetrate no matter what the machine is set to.
What is the difference between arc energy and net heat input?
Arc energy is voltage times current times sixty, divided by travel speed in millimetres per minute times a thousand. Net heat input multiplies that by a process efficiency factor, which runs from 0.6 for TIG up to 1.0 for submerged arc. That is a forty per cent spread, so the same weld can be documented at 1.6 kilojoules per millimetre or 1.0 depending on a convention nobody states. Both figures are shown here for exactly that reason.
Which figure does my welding procedure mean?
Usually net heat input, since most codes write limits against it, but you should check rather than assume. A procedure that says two kilojoules per millimetre maximum without naming the convention permits materially different welds depending on how the reader takes it, and on a TIG procedure the gap between the two readings is the largest of any process. If you are recording heat input for a qualification, record which convention you used alongside the number.
Why does duplex stainless have a minimum heat input?
Because cooling too fast leaves the weld metal with too much ferrite and not enough austenite, which costs both toughness and corrosion resistance. Nearly every other material has only a ceiling, above which the heat-affected zone coarsens or sensitises. Duplex is the one common alloy where welding too cold is a genuine defect rather than merely inefficient, and a thin root pass on heavy section is exactly the situation that produces it.
Why does my stainless weld corrode after welding?
Most often sensitisation. Holding the heat-affected zone in the roughly 450 to 850 degree range lets chromium carbides form at the grain boundaries, which strips chromium from the metal immediately around them and destroys the corrosion resistance exactly there. Keeping heat input down shortens the time spent in that range. Low-carbon grades such as 304L and 316L are far more tolerant of it, which is the whole reason those grades exist.
Is a high deposition rate the same as an efficient process?
No, and confusing them is expensive. Flux-cored wire lays down metal quickly and loses roughly a seventh of it to slag you then chip off. Stick loses nearly two fifths to flux coating and the stub you throw away, so barely three fifths of what you buy reaches the joint. Solid MIG wire is close to ninety-five per cent. Comparing consumables by the kilogram flatters stick enormously unless you account for that.
How accurate is heat input as a guide to weld properties?
It is a proxy rather than a measurement of the thing that matters. What actually decides heat-affected zone properties is the cooling rate, and plate thickness, joint geometry, preheat and interpass temperature all change that without altering the kilojoules per millimetre at all. Two welds recorded at identical heat input on different thicknesses can behave quite differently. Where it genuinely matters the procedure qualification is the authority and a calculation is a planning figure.
🔒 This tool runs entirely in your browser. Nothing you enter is uploaded, logged, or stored.