8 min read

Heat Input and Why It Matters

How heat input is calculated, why travel speed dominates it, the difference from arc energy, and what too much or too little does to a weld.

Heat input is one number that summarises how much energy the arc put into a unit length of joint. It is arithmetic simple enough to do on a scrap of paper, and it controls the cooling rate, the width and structure of the heat-affected zone, the toughness of the finished weld, and how much the assembly moves as it cools. That combination of trivial calculation and serious consequence is why it shows up constantly in examination questions and in procedure review.

This guide covers the calculation and its units, why travel speed carries more weight than most people expect, the distinction between arc energy and heat input, what happens at each end of the range, and how heat input is controlled and verified in real work. It describes general practice; the limits that apply to any real job come from the qualified procedure and the governing code.

The calculation and what it means

Heat input is the arc power divided by the travel speed. Arc power is voltage multiplied by current, which gives watts, or joules per second. Dividing by how fast the arc is moving converts that into joules per unit length, which is the quantity the joint actually experiences. Every version of the formula you will meet is that one idea with unit conversions bolted on to it.

In imperial units the working form multiplies volts by amps by sixty, then divides by travel speed in inches per minute, giving joules per inch. The sixty is only there to reconcile minutes with seconds. In metric the same expression divided by a thousand and by travel speed in millimetres per minute gives kilojoules per millimetre, which is the unit most procedures use.

A worked pair fixes it. At 24 volts and 200 amps moving at 8 inches per minute, the calculation gives 36,000 joules per inch. The same arc moving at 200 millimetres per minute gives 1.44 kilojoules per millimetre. Both describe the same weld, and being able to move between them without hesitation is worth practising.

  • Arc power in watts equals voltage multiplied by current
  • Heat input equals arc power divided by travel speed, expressed per unit length
  • Imperial working form yields joules per inch; metric yields kilojoules per millimetre
  • The sixty in the imperial form is a minutes-to-seconds conversion, nothing more

Travel speed is the variable that moves the number

Voltage and current sit on the top of the expression and travel speed sits on the bottom, which means speed has the most direct leverage. Halving travel speed doubles heat input for exactly the same arc, and doubling it halves the heat input. Nothing else on a welding machine gives that kind of swing for that little effort.

This is why heat input control is really travel speed control in disguise. Voltage on most processes is constrained within a fairly narrow band by arc stability and transfer mode. Current is constrained by the deposition needed and by the electrode. Travel speed is the parameter a welder varies most freely and least consciously, and it is the one that moves the answer most.

It also explains a pattern inspectors see repeatedly. A joint welded by two people to the same nominal parameters can end up with meaningfully different properties because one of them works faster, and the difference will never show up on a machine readout. Only a travel speed measurement catches it.

Arc energy versus heat input

The calculation above strictly gives arc energy, meaning the electrical energy delivered per unit length. Not all of that energy ends up in the workpiece. Some radiates, some is carried away in the plume, some is lost to the electrode and the shielding. Heat input in the stricter sense is arc energy multiplied by a thermal efficiency factor for the process.

The factors differ by process because the physics differs. Submerged arc welding buries the arc under flux and loses very little, so its factor is close to unity. Shielded metal arc, gas metal arc and flux-cored processes sit somewhat lower. Gas tungsten arc welding is the least thermally efficient of the common processes, because a substantial share of the energy goes into the tungsten and the open arc rather than into the joint.

Whether you apply the factor depends on what the governing document asks for, and this is a real source of confusion. Many structural codes work directly in arc energy and call it heat input, while other documents require the efficiency factor. Reading which convention is in use before comparing a calculated value against a limit avoids an entirely avoidable argument.

  • Arc energy - electrical energy per unit length, straight from volts, amps and travel speed
  • Heat input - arc energy adjusted by a process thermal efficiency factor
  • Submerged arc is the most thermally efficient of the common processes
  • Gas tungsten arc is the least, because much of the energy never reaches the joint

What heat input actually controls

The immediate consequence is cooling rate. More energy per unit length means a larger volume of hot metal and a slower cooling curve; less energy means a smaller, faster-quenching weld. Everything else follows from that, which is why heat input and preheat are two levers on the same mechanism and are usually specified together.

Cooling rate sets microstructure. A fast quench favours hard, brittle structures in the heat-affected zone, which is the precondition for hydrogen-assisted cracking. A slow cool favours softer structures but allows grain growth, and coarse grains generally mean lower toughness. Neither end is safe, which is why procedures so often specify a range rather than a ceiling.

Heat input also governs the width of the heat-affected zone and the amount of distortion. A wider zone means more base metal has had its properties altered, and more expansion and contraction means more movement in the assembly and more residual stress locked into it once everything is cold. Distortion control and heat input control are largely the same conversation held in different words.

Too much and too little

Excessive heat input coarsens grain structure in the weld metal and the heat-affected zone, and coarse grains cost impact toughness. It widens the heat-affected zone so a larger volume of base metal has been altered. It produces more distortion, a more fluid puddle that is harder to control out of position, and on some materials it degrades properties the base metal was supplied with. On quenched and tempered and on thermomechanically processed steels, an upper limit is often the more critical of the two.

Insufficient heat input quenches the joint quickly. The heat-affected zone hardens, hydrogen has less time to escape, and cracking risk rises. Very low heat input also raises the practical risk of incomplete fusion, because there may not be enough energy to melt the sidewall properly at the speed being used. Because both extremes cause problems, expect a procedure to specify a band rather than a single ceiling, and treat a one-sided limit as something to ask about.

Measuring it on real work

Voltage and current are the easy half. Both can be read from calibrated instrumentation, and the caution worth knowing is where the voltage is measured: readings taken at the machine include drop through the cables, so a reading at the arc and a reading at the panel are not the same number, and the difference grows with cable length.

Travel speed is the half that gets estimated. The reliable method is to measure a known length of weld and time it, then divide, rather than to ask what speed was used. For mechanised welding the carriage setting can be verified directly, but hand welding needs a stopwatch and a rule, and it is the single most useful measurement an inspector can make when heat input is a controlled variable.

The alternative approach is to work backwards from run-out length: measure how much joint one electrode or one measured length of wire produced, which gives an average travel speed without standing over the welder with a watch. It also smooths out the natural variation in a hand-made weld, which a single timed pass does not.

  1. Confirm which convention the governing document uses: arc energy or efficiency-adjusted heat input
  2. Read voltage as close to the arc as the setup allows and note where it was read
  3. Read current from calibrated instrumentation rather than from a dial setting
  4. Measure a known length of weld and time it to get a real travel speed
  5. Calculate, then compare against the range the procedure permits, not just a maximum
  6. Record the inputs alongside the result so the number can be re-checked later

The caveat on waveform-controlled processes

The simple product of average voltage and average current assumes a reasonably steady arc. Modern pulsed and waveform-controlled power sources deliberately vary both many times per second, and averaging them independently can misrepresent the actual energy delivered, because the peaks in voltage and current are correlated rather than independent.

Where this matters, instantaneous power measurement is the honest approach: sample voltage and current together at high frequency and average the product rather than multiplying the averages. Equipment that reports energy directly is doing exactly this, and where a procedure controls heat input tightly on a pulsed process it is worth knowing which method produced the number in the record.

For most conventional welding the simple calculation is entirely adequate, and it remains the one that examinations ask for. The caveat is worth carrying anyway, because it is the difference between reciting a formula and understanding what the formula is standing in for. Knowing when a shortcut stops being valid is most of what makes it safe to keep using.

Common questions

How is heat input calculated?

Multiply arc voltage by welding current to get arc power in watts, then divide by travel speed to get energy per unit length. In imperial units the working form multiplies volts by amps by sixty and divides by inches per minute, giving joules per inch. In metric it divides by a thousand and by millimetres per minute, giving kilojoules per millimetre.

Why does travel speed have such a large effect?

Because it is the divisor. Halving travel speed doubles heat input for exactly the same voltage and current, and doubling it halves the figure. Voltage and current are usually constrained within narrow bands by arc stability and deposition needs, so travel speed is both the freest variable and the one that moves the result most.

What is the difference between arc energy and heat input?

Arc energy is the electrical energy delivered per unit length, calculated straight from volts, amps and travel speed. Heat input in the stricter sense multiplies that by a thermal efficiency factor for the process, since not all arc energy enters the joint. Many documents use the terms interchangeably, so check which convention applies before comparing a value to a limit.

What happens if heat input is too high?

Grain structure in the weld metal and heat-affected zone coarsens, which typically costs impact toughness. The heat-affected zone widens so more base metal is altered, distortion increases, and on quenched and tempered or thermomechanically processed steels the properties the plate was supplied with can be degraded. This is why an upper limit often matters more than a lower one.

How do I measure travel speed in the field?

Time a known length of weld and divide, rather than accepting a stated speed. For hand welding a rule and a stopwatch are the practical tools. An alternative that smooths out natural variation is to measure how much joint one electrode or one measured length of wire produced, which yields an average travel speed over a realistic run.

Where to go next

More guides