Carbon Equivalent (IIW)

Steel hardens when it cools quickly from above its transformation temperature, and how much it hardens depends on what else is dissolved in the iron. Carbon does most of the work, but manganese, chromium, molybdenum, vanadium, nickel and copper all push in the same direction to varying degrees. The carbon equivalent is the arithmetic that converts each of them into the amount of carbon that would have the same effect, so that one number stands in for a whole certificate. The International Institute of Welding's version — carbon, plus manganese over six, plus chromium, molybdenum and vanadium over five, plus nickel and copper over fifteen — is the one most widely tabulated, and the one most codes still key their preheat requirements to.

Enter the elements as they appear on the mill certificate, in weight per cent. Because the expression is linear and homogeneous, the answer lands on the same scale: a CE of 0.47 means 0.47 % carbon equivalent, which is the number a preheat table is keyed to.

It is a screening number, not a verdict, and the distinction is not pedantry. The expression was fitted to predict hardenability in the heat-affected zone of the steels that were structural practice when it was derived — steels with meaningfully more carbon than a modern grade. On today's low-carbon microalloyed plate it overestimates the risk, sometimes badly. The mechanism is visible in the coefficients: manganese counts a sixth, which is a heavy weighting, and modern grades get their strength from manganese and from microalloy additions rather than from carbon. Work an example. A conventional plate at 0.18 % C and 1.40 % Mn comes out at 0.47. A modern grade at 0.08 % C and 1.50 % Mn comes out at 0.387 — a real improvement, but nothing like the improvement the halved carbon content actually delivers in the plate's cracking behaviour. In the modern steel the manganese term is larger than the carbon term, which is not a description of what makes that steel crack. This is exactly why Ito and Bessyo published $P_{cm}$, and it is why the two parameters can rank the same pair of steels in opposite order.

The preheat comes from a code, not from this page. CE is an input to a table in AWS D1.1, CSA W59, EN 1011-2 or whatever governs your work, and that table also wants the thickness, the restraint and the hydrogen level of the consumable before it will give you a temperature. The table is the authority; this is the number you carry to it. Neither the number nor the table replaces a procedure qualification, which is the only thing that tests your steel, your consumable, your joint and your welder together.

And hydrogen — the other half of cold cracking — appears nowhere in this equation. Hydrogen-assisted cold cracking needs three things simultaneously: a susceptible microstructure, tensile restraint, and diffusible hydrogen. Carbon equivalent speaks to the first. Joint design and fit-up speak to the second. Nothing on this page speaks to the third. Hydrogen comes from moisture and from hydrocarbons: damp flux coating on a rod left out on the bench, a wet or contaminated shielding gas, rust and mill scale, paint, oil, cutting fluid, even the marker used to lay out the joint. Low-hydrogen consumables have to be kept low-hydrogen, which means a sealed tin or a heated rod oven and a rule about how long a rod may sit out. A reader who computes a comfortable CE and stops has addressed one leg of a three-legged stool.

One last practical note: the certificate is for the heat the plate was rolled from, and it reports a ladle analysis. A product analysis taken from the plate itself can differ, and segregation means the composition at mid-thickness is not the composition at the surface — which matters for a full-penetration joint in heavy plate. If the number lands near a threshold in the table, that is not the moment to trust the last decimal place.