When you pull up a preheat requirement for a structural weld, the standard move is Table 5.11—organized by base metal category and plate thickness, it covers most carbon and low-alloy steels in everyday fab work. But Table 5.11 is intentionally conservative. It groups a broad range of chemistries into a single row and sets preheat at the safe high end of that range.
AWS D1.1:2025 Annex B exists for situations where that conservatism costs you something—time, heat distortion risk on thin material, or coverage of a base metal the table simply doesn't list. Annex B provides a chemistry-based alternative: calculate a carbon equivalent (CE) from the actual mill test report and use that number to determine minimum preheat.
Rule library based on AWS D1.1:2025; verify against your governing edition.
Why Preheat Matters in the First Place
Before walking through Annex B, it helps to understand what preheat is actually doing. Hydrogen-induced cracking (HIC)—sometimes called cold cracking or underbead cracking—is the primary threat preheat addresses. When welding low-alloy steels, the HAZ can transform to hard martensite as the weld cools. If diffusible hydrogen (from moisture, atmospheric contamination, or electrode coating) is also present, hydrogen can migrate to the high-stress, high-hardness HAZ and initiate cracks hours or even days after the arc goes out.
Preheat slows the cooling rate, which limits martensite formation and gives hydrogen more time to diffuse out of the joint harmlessly. The higher the base metal's hardenability—its tendency to form martensite—the more heat you need to slow that cooling rate adequately.
Hardenability is driven primarily by chemistry. Carbon is the strongest hardenability element, but manganese, chromium, molybdenum, vanadium, nickel, and copper all contribute. The carbon equivalent distills all of those contributions into a single scalar that correlates with cracking susceptibility.
What Annex B Provides
Annex B in AWS D1.1:2025 is an informative annex—meaning it provides guidance and a methodology, but it does not replace the normative requirements of Table 5.11 for purposes of establishing a prequalified WPS. It gives you a CE formula and a framework for translating that CE into a minimum preheat temperature, accounting for joint restraint and base metal thickness.
The key departure from Table 5.11 is that Annex B anchors preheat to the actual heat analysis from the MTR rather than to a category that may lump together steels ranging from lean A36 chemistry to fully alloyed A588. When a mill produces an A572 Grade 50 plate to a low-CE practice, Annex B may show that the conservative Table 5.11 value for that material category is not strictly necessary.
Using the Carbon Equivalent Calculation
To apply the Annex B method, you need the actual heat chemistry for every base metal included in the weld connection. Pull the MTR for every heat of steel going into the joint—not the ASTM specified minimums, the actual tested values. The CE formula in Annex B accounts for carbon plus weighted contributions from the principal alloying elements. Heavier alloying and higher carbon both drive CE up.
Once you have the CE for the base metals (and, where relevant, the deposited weld metal chemistry), Annex B provides guidance on minimum preheat as a function of CE, plate thickness, and joint restraint category. A low-restraint single-pass fillet weld on thin plate with clean chemistry may show no preheat required. A high-restraint thick CJP groove weld on moderately alloyed material will still need meaningful preheat, and the Annex B result should track closely with Table 5.11.
When you have a dissimilar-metal joint—two different base metals meeting at a groove weld—use the higher of the two CE values to drive preheat selection. The more hardenable side dictates the minimum.