Skip to content
Home Articles Standards About Contact
Buying Guide Field Study

Chromium Molybdenum Steel Welding: The Complete Guide

NH
Nicholas Harper · Dec 28, 2025 · 41 min read
Table of Contents

A refinery crew completes a 2.25Cr-1Mo pipe weld at 11 p.m., the rain starts, nobody covers the joint, and the weld goes cold overnight. Two days later a hairline crack appears under the cap, inches from a weld that looked perfect on radiography. Chromium molybdenum steel almost never fails loudly — it fails slowly, invisibly, and expensively, which is exactly why this guide exists.

Chromium Molybdenum Steel Welding: Why This Alloy Family Demands a Different Playbook

A refinery crew finishes a 2.25Cr-1Mo pipe weld at 11 p.m. Rain starts, nobody tarpaulins the joint, and the weld cools uncontrolled overnight. Two days later a hairline crack shows up under the cap, inches from a weld that passed visual inspection and looked perfect on film. Chromium molybdenum steel rarely fails loudly. It fails quietly, invisibly, and expensively — and that is exactly why this guide exists.

What Chromium Molybdenum Steel Welding Actually Means

Chromium molybdenum steel — usually shortened to Cr-Mo or chromoly — is plain carbon steel deliberately alloyed with chromium and molybdenum. Chromium buys oxidation and corrosion resistance and boosts hardenability; molybdenum raises high-temperature creep strength and counters temper embrittlement. Those same two elements are precisely what make welding tricky: they make the heat-affected zone (HAZ) form hard martensite very readily, even at modest cooling rates.

So can Cr-Mo steel be welded reliably? Yes. It is welded every day in refineries, power stations, and aircraft hangars. But only under three non-negotiable conditions: controlled preheat, genuinely low-hydrogen practice, and — for most creep-resistant grades — post-weld heat treatment (PWHT). Treat those as suggestions and the joint will eventually crack, often days after everyone has gone home.

Where You Will Meet This Material

The applications are high-stakes by definition, because Cr-Mo steel is specified when carbon steel would creep, scale, or soften in service:

  • Power generation — boiler tubes, main steam and hot reheat piping, turbine casings running for decades at 500–600°C.
  • Petrochemical and chemical plants — hydroprocessing reactor circuits, crackers, and refinery piping where hydrogen and sulfur are present.
  • Pressure vessels — ASME-stamped reactors, drums, and heat exchangers built to Section VIII.
  • Aerospace and motorsport — 4130 and 4140 tube frames, roll cages, landing gear, shafts, and suspension components where strength-to-weight matters. If you work in this space, our Aerospace Welding: Processes, Standards, and Careers Guide covers the broader fabrication environment.
  • Oilfield equipment — drillstring components, downhole tooling, and wellhead hardware under API oversight.

Why Carbon-Steel Habits Fail Here

A welder who runs 7018 on A106-B pipe all week develops habits that are quietly lethal on P22. On carbon steel, a little surface rust is tolerable, a cold start is survivable, and air cooling after the last pass is normal. On 2.25Cr-1Mo, hydrogen picked up from a damp covering, a hard untempered HAZ, and residual stress combine into the classic cold-cracking triad. The weld looks fine. The crack comes later.

Note: AWS D10.8-96, Recommended Practices for Welding of Chromium-Molybdenum Steel Piping and Tubing, covers filler metal, joint design, preheating and postheating, with special emphasis on interpass temperature control and the real dangers of interrupted heating cycles — the exact failure mode in the story above. See the standard on the AWS publishing site.

Cr-Mo steels are used extensively in the power, chemical, and petroleum industries, and their fabrication is governed by a stacked framework of codes: ASME B31.1 (power piping), B31.3 (process piping), the Boiler and Pressure Vessel Code Sections I, IIA, IIC, VIII and IX, plus API 570 for in-service piping inspection and repair and the National Board Inspection Code (NBIC) for repairs and alterations. The appropriate code depends entirely on what you are building and where it will run.

What This Guide Covers

The rest of this article is built to take you from theory to a weld that survives inspection and service. You will get:

  1. A clear map of the Cr-Mo families, from AISI 4130 chromoly to Grade 91, with compositions and P-Numbers.
  2. The metallurgy of cracking — hydrogen, reheat cracking, temper embrittlement, and Type IV — in plain language.
  3. Filler metal selection logic, including when undermatching beats matching. Our Alloy Steel Welding Electrodes: Selection, Use & Safety guide supplements the consumable basics here.
  4. Practical preheat and interpass tables you can take straight to the joint.
  5. PWHT rules per ASME, including the Grade 91 Ac1 limit.
  6. Real-world procedures for 4130, 4140, and Cr-Mo power piping.
  7. Temper bead and controlled-fill alternatives when PWHT is impossible.
  8. Crack diagnosis, inspection discipline, and a jobsite checklist.

Before the numbers start, it helps to understand what the alloy itself is doing during welding — because once you can picture martensite forming under your bead, every preheat and soak time on the following pages stops feeling like bureaucratic ritual and starts making physical sense. ESAB’s Cr-Mo welding guidelines are a useful companion reference as you work through the material.

Know Your Grade: The Cr-Mo Steel Families and What Each One Changes

“Chromium molybdenum steel” is a label broad enough to cover a motorcycle roll cage and a refinery’s main steam line. Those two weldments share an element list and almost nothing else. Before you pick a filler or set a preheat, you need to know which of the two Cr-Mo families you are actually holding, because the grade drives everything downstream: hardenability, cracking risk, heat treatment, code requirements, and how much trouble a bad day costs you.

Family One: The Heat-Treatable Chromoly Grades

The first family is the low-alloy, quench-and-temper group that most fabricators meet as bar, tube, and plate: AISI/SAE 4130, 4140, and 4340, universally called “chromoly.” These are hardenable steels whose strength comes from a heat-treatment cycle, not from creep resistance. You will find them in aircraft structures, motorsport chassis and roll cages, bicycle frames, drive shafts, connecting rods, gears, and tooling.

  • 4130 carries 0.28–0.33% carbon with roughly 1% chromium and 0.20% molybdenum, giving a carbon equivalent near 0.60. It is the weldable one of the bunch, strong yet forgiving enough for thin-wall tube work.
  • 4140 pushes carbon to about 0.40% with similar chromium and molybdenum. That extra carbon buys tensile strength and simultaneously buys hardenability, so the heat-affected zone runs much harder and cracks far more readily.
  • 4340 adds 1.65–2.00% nickel on top of the 4140 recipe, which dramatically increases through-thickness hardenability. It behaves like 4140 with a longer fuse.

Typical products include ASTM A519 mechanical tubing, 4140/4130 bar and plate, and seamless tube. The temper condition matters as much as the chemistry: annealed or normalized 4140 welds respectably, while 4140 already hardened above roughly 25 Rc is a cracking problem waiting for a hydrogen source. For a deeper look at consumables for this group, see our guide to alloy steel welding electrodes.

Family Two: The Creep-Resistant Power and Petrochemical Grades

The second family is built for high-temperature service, where the enemy is not impact loading but slow, time-dependent creep. These steels are grouped by ASME P-Number because pressure-vessel and piping codes qualify procedures by P-Number rather than by trade name. ESAB’s Cr-Mo welding guidance lists P1 (C-0.5Mo), P11 (1.25Cr-0.5Mo), P22 (2.25Cr-1Mo), P5 (5Cr-0.5Mo), P9 (9Cr-1Mo), and P91 (9Cr-1MoVNb) as the standard piping grades.

Grade ASME P-No. Nominal composition Typical products Typical service limit
C-0.5Mo P1 0.5Mo A335 P1 pipe, A387 Gr 1 plate ~480 °C (900 °F)
1.25Cr-0.5Mo P11 1.25Cr / 0.5Mo A335 P11 pipe, A387 Gr 11 plate ~540 °C (1000 °F)
2.25Cr-1Mo P22 2.25Cr / 1Mo A335 P22 pipe, A387 Gr 22 plate ~565 °C (1050 °F)
5Cr-0.5Mo P5 5Cr / 0.5Mo A335 P5 pipe, A387 Gr 5 plate ~595 °C (1100 °F)
9Cr-1Mo P9 9Cr / 1Mo A335 P9 pipe, A387 Gr 9 plate ~595 °C (1100 °F)
Grade 91 (9Cr-1Mo-V-Nb-N) P15E (P91) 9Cr / 1Mo / V / Nb / N A335 P91 pipe, A387 Gr 91 plate, A213 T91 tube up to about 600 °C

The service temperatures above are typical design envelopes, not approval limits; always confirm against the governing code and the equipment data sheet. Note the trend: as chromium climbs, so does hardenability and so does the temperature at which the alloy keeps its strength. Grade 91 is the endpoint of that climb, a creep-strength-enhanced ferritic steel whose load-bearing capability depends on a precise tempered martensite microstructure reinforced by controlled vanadium-nitrogen and niobium-carbon precipitation. Disturb that precipitation with the wrong heat treatment and you do not get “slightly weaker” steel — you get a grade that no longer performs as Grade 91. This is why power-piping work sits under codes such as AWS D10.8, Welding of Chromium-Molybdenum Steel Piping and Tubing.

Short Note: Many fabricators try to estimate preheat from the IIW carbon equivalent formula, CE = C + Mn/6 + (Cr+Mo+V)/5 + (Ni+Cu)/6. ESAB states plainly that the IIW carbon equivalent method is not valid for these Cr-Mo grades, because chromium and molybdenum contribute to hardenability far differently than the formula assumes, and higher chromium also changes how hydrogen diffuses out of the joint. Do not let a low calculated CE talk you out of preheat. For Cr-Mo, follow the code table or the consumable manufacturer’s written recommendation instead.

The practical takeaway is simple: identify the grade first, in writing, before anything gets struck. Confirm the alloy and its delivered condition — annealed, normalized, normalized-and-tempered — because the same P22 pipe welds very differently depending on how it arrived, and because aerospace work on chromoly tube follows its own expectations, covered in our overview of aerospace welding standards. Once the grade and condition are fixed, every other decision in this guide has a defensible answer.

Why Cr-Mo Welds Crack: Hydrogen, Martensite, Reheat Cracking and Temper Embrittlement

Chromium-molybdenum welds rarely fail because someone could not run a bead. They fail because of what happened inside the metal after the arc went out. Four mechanisms account for nearly all cracking in this alloy family, each with a different trigger, timescale, and fix: hydrogen-induced cold cracking, hot and reheat cracking, temper embrittlement, and — in Grade 91 — Type IV cracking.

Hydrogen-Induced Cold Cracking: The Three-Way Coincidence

The most common Cr-Mo failure is hydrogen-induced cold cracking, also called delayed or underbead cracking. It forms when three conditions exist at the same time in the same place:

  • Diffusible hydrogen in the weld metal and HAZ, mostly from moisture in coatings, slag, rust, oil, or ambient humidity.
  • A hard, susceptible microstructure, meaning martensite or a high-carbon bainite/martensite mix.
  • Tensile stress from weld shrinkage, joint restraint, and geometric stress raisers.

Remove any one of the three and the crack does not form. That single fact is the entire logic behind preheat, low-hydrogen consumables, and sensible joint design — they are not three separate rules, they are three ways of attacking one triangle.

Here is why Cr-Mo is so much more sensitive than carbon steel. Above the upper critical temperature (Ac3), the HAZ briefly becomes austenite, a phase that holds far more carbon in solution than ferrite. As chromium and molybdenum content climbs, the continuous cooling transformation curve shifts right, so even moderate cooling rates produce martensite instead of soft ferrite and pearlite. Hardenability climbs steeply with grade: 9Cr material can form martensite on air cooling in heavy sections. Fresh martensite is hard, brittle, and supersaturated with hydrogen, which migrates toward stress concentrators — prior austenite grain boundaries, inclusion interfaces, notch roots — recombines into molecular hydrogen, and builds local pressure until the metal separates. The AWS D10.8 practice for welding Cr-Mo piping and tubing is built around exactly this problem, and ESAB’s Cr-Mo welding guidelines walk through the same hydrogen-control framework in shop-floor language.

Note: Cold cracking is delayed by nature. Hydrogen needs time to diffuse, so a weld that looks perfect at the end of the shift can develop a hairline under-bead crack hours or even days later. Holding a weld before releasing restraint, applying a hydrogen post-heat, and scheduling crack-sensitive NDE properly are all part of the job. More on inspection timing in Section 9.

Hot and Reheat Cracking: The J-Factor and Impurity Control

Hot cracking (solidification cracking) and reheat cracking (stress-relief cracking) are a different family, driven by impurity elements — phosphorus, tin, sulfur, antimony, arsenic and copper — that segregate to grain boundaries and weaken them. Reheat cracking typically nucleates in the coarse-grained HAZ during PWHT or high-temperature service, where creep-strengthening precipitates form faster than the grain boundaries can relieve stress by sliding, so voids open along the boundaries.

The industry quantifies this susceptibility with the Watanabe J-factor:

J = (%Si + %Mn)(%P + %Sn) × 10⁴

The practical target is J below 150, with copper held to a maximum of 0.20% and nickel to 0.30% — the chemistry window set by API 934 for 2.25Cr-1Mo-V plate and pipe in hydroprocessing service. These are purchasing-specification controls: once a heat of steel is melted with too much tin and phosphorus, no amount of welding technique fixes the base metal. If you are ordering Cr-Mo material for long-term high-temperature service, J-factor belongs on the purchase order.

Temper Embrittlement: The Repair PWHT Trap

Extended exposure in roughly the 700–1100°F (370–595°C) range drives phosphorus, tin, antimony and arsenic to prior austenite grain boundaries, dropping room-temperature toughness without any visible change. After years in service, most Cr-Mo pressure equipment is already partially embrittled — it simply has not been asked to absorb an impact at low temperature yet. The trap is that repair PWHT often spends hours right inside that danger range while ramping up and down slowly. The mitigation is chemistry control again (the related X-bar impurity parameter alongside J-factor), a fast enough ramp to pass through the range quickly, and a realistic assessment that service-exposed metal is not the same metal that went into the vessel decades ago.

Type IV Cracking: Grade 91’s Signature Failure

Grade 91 has its own characteristic creep failure mode. Type IV cracking occurs preferentially in the fine-grained HAZ immediately adjacent to the intercritical HAZ — the narrow band reheated just above Ac1 but below Ac3. That band is partially transformed and, after PWHT, ends up as the weakest zone in the weldment: fine-grained ferrite and freshly tempered martensite with coarsened carbides and MX carbonitrides that have lost their strengthening effect. Creep voids nucleate there in service, and the crack runs parallel to the weld in a thin, easily missed band.

What makes this urgent is how easily that weak zone is created by accident. A study of simulated P91 HAZ subzones found the coarse-grained HAZ needed 8 hours at 740°C, 4 hours at 760°C, more than 1 hour at 780°C, or just 0.5 hours at 800°C to reach a properly tempered 265 HV. The common 3-hour-at-760°C PWHT can therefore leave the coarse-grained zone under-tempered and hard. EPRI has documented Grade 91 components already being replaced or repaired because improper heat treatment produced properties below code minimums — failures baked in on day one that surfaced as creep cracks years later.

Quick Note: If you are welding or repairing a Cr-Mo component and the failure was a slow, tight crack rather than a gross overload break, start with hydrogen (if it appeared within a few days), then reheat cracking (if it appeared during or after PWHT), then Type IV (if it is Grade 91 in service). Diagnosing the mechanism correctly determines whether the fix is a drying oven, a new WPS, or a new heat of plate.

The through-line in all four mechanisms is that the danger is metallurgical, not cosmetic. Bead appearance tells you almost nothing about hydrogen content, grain-boundary chemistry, or whether a fine-grained band 2 mm from the fusion line is about to become a creep crack. That is why chemistry limits, thermal control, and disciplined heat treatment — not welding skill alone — define whether a Cr-Mo weld survives its service life. The next two sections turn that metallurgy into the numbers you actually use: filler selection, preheat, and post-weld heat treatment. Alloy steel welding electrodes covers consumable handling in more depth if you need it.

Filler Metal Selection: Matching, Undermatching and the AWS Classification System

Filler choice is where a lot of Cr-Mo jobs quietly go wrong. The arc looks fine, the bead stacks nicely, and nobody checks whether the wire actually belongs on that base metal. In practice, filler selection comes down to three questions: what condition is the base metal delivered in, what condition must the joint be in after welding, and what does the design actually require of the weld? Hobart Brothers’ technical guidance puts it bluntly: filler choice depends on base-metal condition, required post-weld condition and design requirements, and overmatching tensile strength is generally not recommended because it reduces ductility and increases crack susceptibility.

SMAW Electrodes: The AWS A5.5 Lineup

Stick electrodes for low-alloy Cr-Mo steels are classified under the AWS A5.5 specification. The suffix tells you the alloy family: B1 is 0.5Cr-0.5Mo, B2 is 1.25Cr-0.5Mo, B3 is 2.25Cr-1Mo, B8 is 9Cr-1Mo, and B9 is the vanadium-modified Grade 91 matching filler. The most common selections look like this:

AWS class Matches Typical use
E7018 Undermatched to most Cr-Mo Repair, soft fill, 4130/4140 where Q&T isn’t planned
E8018-B1 0.5Cr-0.5Mo Light elevated-temperature service
E8018-B2 1.25Cr-0.5Mo (P11) Refinery and boiler piping
E9018-B3 2.25Cr-1Mo (P22) Heavy-wall pressure vessels and power piping
E8018-B8 9Cr-1Mo (P9) High-temperature steam service
E9018-B9 Grade 91 Creep-strength-enhanced ferritic piping

A useful shortcut exists for shops that want to skip procedure development: AWS publishes a Standard Welding Procedure Specification for SMAW of chromium-molybdenum steel. AWS B2.1-5A-223 covers base metal M-5A/P-5A using E9018-B3, and is valid from 1/8 to 1/2 in. thick in the as-welded condition, and 1/8 to 1-1/2 in. in the PWHT condition. If your joint falls inside those ranges, an approved SWPS can save you a full qualification round. For broader background on reading electrode classifications, see our guide to alloy steel welding electrodes.

GMAW and GTAW Wires: AWS A5.28

For wire processes, the AWS A5.28 classification system works the same way. Common choices are ER70S-2 (mild, ductile, general-purpose), ER80S-D2 (a step up in strength, popular on 4140), ER80S-B2 and ER80S-B3 for 1.25Cr and 2.25Cr work, ER90S-B3 for heavier P22 joints, and ER80S-B8 or ER90S-B9 for 9Cr and Grade 91. ER4130 exists specifically for matching 4130 weldments, but it comes with a caveat we cover below.

Quick Note: A GMAW study on AISI 4140 using ER4130 filler found the wire highly susceptible to porosity, making it difficult for full-penetration butt joints. If you choose ER4130, rigorously control shielding gas flow, joint cleanliness and drafting, or consider ER80S-D2 instead for critical full-penetration work.

Matching, Undermatching and Why Overmatching Is Discouraged

Matching filler means the deposited metal closely follows the base metal’s nominal composition and strength. That is the right call when the entire weldment will be quenched and tempered afterward, because filler and base metal then respond to heat treatment as one system. Outside that scenario, things shift.

Undermatching deliberately uses a softer, more ductile filler than the base metal. Roll cages, chassis, suspension components and repair welds commonly use E7018 or ER70S-2 on 4130 and 4140 for good reasons: the weld metal gains ductility and fatigue resistance, and the joint tolerates misalignment and residual stress better. Overmatching, by contrast, is generally discouraged. A harder, stronger bead sits next to a heat-affected zone that is already the weakest, most brittle link, and it tends to push cracking into that HAZ rather than giving it anywhere harmless to go.

Hydrogen Ratings: H4, H8, H10 and Storage Rules

Low-hydrogen consumables carry a diffusible hydrogen suffix: H4 means a maximum of 4 ml per 100 g of deposited weld metal, H8 means 8 ml, H10 means 10 ml. Lower is better, and for Cr-Mo work you generally want the lowest you can get. For Grade 91, EPRI specifies that SMAW electrodes be certified to the H4 designation and issued from a heated master storage bin, which means rod ovens, not cartons left on the bench. Electrodes that have picked up moisture need rebaking per the manufacturer’s schedule, and exposed low-hydrogen rods should never be left overnight without heat. Position-specific consumables matter too, so check our all-position welding rod guide before assuming a flat-only electrode will behave overhead. AWS D10.8 also frames filler and procedure requirements for Cr-Mo piping and tubing (see the standard).

Preheat and Interpass Temperature: The Practical Numbers Table

For chromium molybdenum steel welding, temperature control is less of a precaution and more of a safeguard against delayed cracking. The addition of chromium and molybdenum greatly increases hardenability, meaning the heat-affected zone can form martensite at relatively modest cooling rates. Preheat slows that cooling, allowing dissolved hydrogen to diffuse out before the steel hardens. It also reduces the thermal gradient between the weld and the cold base metal, limiting both residual stress and the depth of the fully hardened zone.

The Practical Numbers: Preheat and Interpass by Grade

For creep-resistant Cr-Mo piping, ESAB, Guidelines for Welding Chromium Molybdenum Steel provides the following minimums for low-hydrogen SMAW or GMAW. A qualified WPS may require higher values:

Grade (P-Number) Thickness Minimum preheat Maximum interpass
P11 (1.25Cr-0.5Mo) Up to 15 mm (~0.6 in) 100 °C (210 °F) 300 °C (570 °F)
P11 Above 15 mm 150 °C (300 °F) 300 °C (570 °F)
P22 (2.25Cr-1Mo) Up to 15 mm 150 °C (300 °F) 350 °C (660 °F)
P22 Above 15 mm 200 °C (400 °F) 350 °C (660 °F)
P5 (5Cr-0.5Mo) All thicknesses 200 °C (400 °F) Per qualified procedure
P91 (9Cr-1Mo-V-N) All thicknesses 200 °C (400 °F) 315 °C (600 °F)

The trend is clear. As chromium and molybdenum increase, so does the required floor. For P5 and P91, the 200 °C (400 °F) minimum is applied to all thicknesses due to their high hardenability.

For the heat-treatable chromoly grades, Hobart Brothers’ guidance on 4130 and 4140 takes a thickness-based approach:

  • Under 3/16 in (~4.8 mm): generally no preheat required with low-hydrogen consumables.
  • 3/16 to 1/2 in: 300–400 °F (150–200 °C).
  • 1/2 to 1 in: 400–500 °F (205–260 °C).
  • Over 1 in: 600–800 °F (315–425 °C).

This ladder is especially relevant for aerospace TIG work, where thin-wall 4130 can dissipate heat rapidly. Heavier 4140 forgings and shafts, however, have a strict ceiling: the interpass temperature must never exceed the base metal’s original tempering temperature, typically well under 600 °F for as-quenched-and-tempered stock. Exceeding it can re-temper the parent metal and undo its delivered strength and hardness.

Quick Note: If the joint falls below the minimum preheat before completion, stop welding. Re-preheat, re-clean the surface, and resume. Hydrogen already diffused into a completed weld cannot be driven out by simply adding another bead. That requires a documented hydrogen bake-out.

Grade 91: Tight Upper Limits

Grade 91 calls for a 200 °C (400 °F) minimum preheat and a 315 °C (600 °F) maximum interpass. The upper limit is critical, as exceeding it can soften the fine-grained HAZ and contribute to Type IV weakness. EPRI repair research found that repairs at 149 °C (300 °F) preheat with a 290 °C (550 °F) maximum interpass showed no hydrogen-induced cracking when strict low-hydrogen consumable handling was enforced (EPRI). This reinforces that preheat and hydrogen control must work together.

Extent, Measurement, and Hydrogen Control

Preheat must extend a minimum of 3 in (75 mm) in every direction from the joint, through the full thickness, and remain uniform while welding. As outlined in AWS D10.8, failing to heat a wide enough area can create a hard zone just outside the preheated band, an area prone to cracking.

To stay compliant, measure rather than guess:

  • Contact pyrometer: The most reliable method, taken directly on the preheated surface away from the heating source.
  • Tempilstik or temperature crayons: Useful for quick field checks by bracketing the target temperature.
  • Infrared pyrometer: Fast, but requires emissivity correction. Mill scale, oxide, or ground surfaces can be misleading. As noted by the AWS Welding Digest’s preheat and interpass review, verify IR readings against a contact measurement at least once per job.

Record preheat and interpass readings pass by pass. Interpass must never fall below the preheat minimum or exceed the grade ceiling, as consecutive hot passes can drift out of range quickly. If post-weld heat treatment will be delayed, hold the joint at preheat temperature (commonly 300–400 °F) immediately after welding as a temporary hydrogen bake-out until PWHT can be performed. Lastly, only use dry, properly stored alloy steel welding electrodes, such as E7018, E8018-B2, and E9018-B3. Even perfect preheat cannot compensate for damp consumables.

Post-Weld Heat Treatment: Codes, Temperatures, Hold Times and the A1 Limit

If preheat is the first line of defense in chromium molybdenum steel welding, post-weld heat treatment (PWHT) is the second, and in heavy-section creep-resistant grades it is the one that actually decides whether the joint survives twenty years of service. PWHT does three jobs at once: it tempers the hard martensite formed in the weld metal and heat-affected zone, it reduces the residual stresses left by restrained contraction, and it drives out diffusible hydrogen that preheat alone may not have released. Skip it on a P22 header, or botch the temperature window, and you can undo every good habit that came before.

The ASME Framework: UCS-56 Tables by P-Number

ASME BPVC Section II Part D and Section VIII Division 1 UCS-56 provide the code minimums, organized by P-Number rather than by trade name. For P-Number 4 steels (2.25Cr-1Mo, P22), the code minimum PWHT temperature is 1200 °F (650 °C); for P-Number 5A, 5B and 5C steels (5Cr-0.5Mo and the 9Cr family) it rises to 1250 °F (675 °C). The standard holding time is one hour per inch of thickness, with a minimum usually applied for thin walls. UCS-56 does grant limited exemptions from mandatory PWHT for thin-wall pipe at moderate strength levels, but exemptions are thickness- and preheat-condition-dependent, and most pipeline and refinery work under boiler and pressure-vessel practice still specifies it.

Widely used shop ranges, per ESAB’s Cr-Mo welding guidance, look like this:

Grade Typical PWHT Range Minimum Hold
P11 (1.25Cr-0.5Mo) 650–700 °C (1200–1290 °F) 1 h per 25 mm (1 in.)
P22 (2.25Cr-1Mo) 680–720 °C (1255–1330 °F) 2 h minimum
P5, P9, P91 730–760 °C (1345–1400 °F) 2 h minimum

Heating and Cooling Rates: D1.1 Clause 7.8 and Local PWHT

Temperature is only half the story. Ramping too fast reintroduces the thermal stress PWHT is meant to relieve. AWS D1.1 Clause 7.8 sets the familiar limits: entry temperature no higher than 800 °F, heating rate no faster than 400 °F per hour divided by the maximum thickness in inches, cooling no faster than 500 °F per hour divided by that thickness, and below 800 °F the joint may cool in still air. A 4-inch header, for example, caps heating at 100 °F/hr and cooling at 125 °F/hr.

When a furnace is not an option, local (in-situ) PWHT of Cr-Mo piping and tubing is governed by AWS D10.8, which defines the required temperature uniformity, thermocouple placement and soak band width around the weld. AWS D10.10 covers the general requirements for local heat treatment of weldments, and both assume the same controlled ramp-and-soak philosophy. The governing principle is that the whole joint, plus a generous band either side, must see the specified temperature — not just the cap.

Grade 91 and the Ac1 Ceiling

Grade 91 (P-Number 15E) is where PWHT gets genuinely dangerous. ASME sets a 1300 °F (705 °C) minimum, but industry best practice, guided by EPRI’s Grade 91 work, narrows the window to 1350–1425 °F (730–775 °C) with a minimum two-hour hold. The upper bound matters as much as the lower one: Grade 91’s lower critical temperature (Ac1) sits between 788 °C and 813 °C (about 1450–1495 °F). Cross it and you begin re-austenitizing the HAZ, so on cooling you form fresh martensite plus delta ferrite in a joint that has already been through its final temper — the exact microstructure PWHT was supposed to eliminate. The practical rule is to stay roughly 30 °C below Ac1. When nickel plus manganese content is 1.0% or less, a maximum of 1455 °F is generally accepted; higher Mn+Ni depresses Ac1 further and tightens that ceiling.

Note: PWHT should follow welding promptly. EPRI recommends that Grade 91 joints enter the furnace within 8 hours of completing the weld, or receive an immediate hydrogen post-heat. Delay allows hydrogen to migrate to hard zones and crack to initiate before any tempering occurs.

PWHT for 4130 and 4140: Don’t Over-Temper

Heat-treatable chromoly is a different problem. These alloys are usually supplied quenched and tempered to a specific hardness, and PWHT at power-piping temperatures would simply soften them. The correct approach is to temper at roughly 50 °F below the original tempering temperature, which relieves stress without significantly sacrificing strength. If the weldment distortion is unacceptable or hardness is out of spec, the alternative is a full re-quench and temper of the entire part. Forgetting the original temper and guessing is how roll cages and shafts end up 15 Rc softer than designed, as discussed in our look at alloy steel welding electrodes and their matching heat treatment.

Done right, PWHT is unglamorous and completely routine: correct window, controlled ramp, verified soak, documented chart. Done wrong — overshoot past Ac1, rushed ramps, uneven local heating — it produces cracks that look exactly like welding defects and get blamed on the welder. Treat the furnace chart with the same seriousness as the WPS.

Process and Procedure: Welding 4130, 4140 and Cr-Mo Pipe in the Real World

Everything up to this point is theory until it meets an actual joint. In practice, Cr-Mo work falls into three very different buckets: thin-wall 4130 fabrications, heavy 4140 machined parts, and power piping. Each has its own habits, and mixing them up is how welds get rejected.

Thin-Wall 4130: Aircraft, Cages and Bicycle Frames

Chromoly tubing in the 0.035–0.083 in wall range is forgiving stuff when it arrives soft, which is why it dominates aerospace TIG welding work, roll cages, and bike frames. The standard play is GTAW with ER70S-2 or ER80S-D2 filler: both are undermatching, low-hydrogen options that deposit a ductile weld bead in a joint that will never see the heat treatment needed to make a matching filler worthwhile. Matched ER4130 wire is reserved for weldments that will be fully re-quenched and tempered afterward.

  • Fit-up is the whole job. Tight, square, gap-free joints keep heat input low and prevent burn-through on thin wall. Tack in four places and check alignment before the first pass.
  • Preheat is usually minimal or none on thin, low-carbon-equivalent tubing in the annealed or normalized state. Anything that does need preheat gets it, verified with a contact pyrometer or Tempilstik rather than judged by color.
  • Air cooling is acceptable for thin-wall 4130 in the soft condition because the HAZ stays small and the section cannot trap much hydrogen.
  • Back purge the root on tubing where the inside will not be reamed, and grind the start/stop tapers so fatigue cracks have nowhere to begin.

4140: Annealed or Normalized Only

Heavy 4140 is a different animal, and the first question is always the delivered condition. Hobart’s guidance is blunt: weld 4130 and 4140 only in the annealed or normalized condition. A general shop rule of thumb is that anything over 25 Rockwell C is difficult to weld, and through-hardened material is asking for trouble, because the HAZ will quench to untempered martensite the moment the arc moves on.

Practical sequence for a 4140 part:

  1. Confirm hardness with a portable tester, not a guess. Above 25 Rc, stop and soften the part first.
  2. Preheat per the procedure, then weld with complete joint continuity — no cold starts, no skipped areas, no stopping mid-run where a start/stop defect can sit in a hard zone.
  3. Cool slowly under insulation, or apply a hydrogen post-heat immediately after welding if PWHT will be delayed.
  4. Temper about 50 °F below the original temper, or re-quench and temper the whole part if dimensional stability allows.

Qualification matters here: 4140 is not a prequalified base metal under AWS D1.1. There is no shortcut WPS to pull off the shelf. The procedure must be qualified by testing per AWS B2.1 or ASME Boiler and Pressure Vessel Code Section IX, with the essential variables — filler classification, preheat, heat input, post-heat — written into the qualification record.

Power Piping: Root Pass Control and the No-Cold-Joint Rule

On ASTM A335 P11, P22, P5, P9 or P91 pipe, the procedure is built around the root. Use a proper fit-up with a backing ring or consumable insert where the code allows it, keep root heat input low enough to avoid burn-through, and get full penetration on the first pass because you will not be able to fix a lack of fusion from underneath later. AWS D10.8 covers welding of chromium-molybdenum steel piping and tubing and places particular emphasis on maintaining interpass temperature and on the dangers of interrupted heating cycles.

The critical field rule is simple: the weld never goes cold mid-job. If a joint drops below the specified preheat temperature — a break, a shift change, rain on an uninsulated joint — welding stops. Per D10.8, the joint must be re-preheated and re-inspected before work resumes. A hairline underbead crack from an uncontrolled cooldown will still read clean on a hurried visual.

Note: Dissimilar joints are not improvisation territory. AWS D10.8 explicitly covers welding similar and dissimilar Cr-Mo steels, Cr-Mo to austenitic chromium-nickel stainless, and Cr-Mo to carbon or carbon-molybdenum steels. On a Cr-Mo to stainless transition, use a nickel-base or higher-alloy filler, keep preheat on the Cr-Mo side only, and remember that carbon migration across the fusion line is why these joints are managed with a documented procedure rather than habit.

Whichever bucket you are in, the pattern repeats: correct delivered condition, correct filler, verified heat, uninterrupted execution. That consistency is what separates a Cr-Mo weld that lasts from one that cracks quietly a few days later.

When You Cannot PWHT: Temper Bead, Controlled Fill and Alternative Strategies

Real jobsites rarely hand you a furnace. A nozzle is welded into an operating vessel, a cracked fitting sits six stories up, or a power-piping tie-in cannot be baked because the surrounding equipment will not tolerate it. When full post-weld heat treatment is off the table, you need a documented fallback — and the welding codes have one: temper bead techniques, controlled-fill repair, and carefully chosen “cold” fillers.

The Metallurgical Trick: Let the Next Bead Temper the Last One

Temper bead (sometimes called temper layer) welding exploits a simple idea. Every pass you deposit reheats the heat-affected zone underneath it. If you control bead size, placement and heat input, each new pass softens the hard martensite formed by the previous pass — essentially performing localized PWHT one layer at a time, without ever taking the joint above the code’s temperature limits.

The results are measurable. In one study of AISI 4140 steel welded by GMAW, conventional as-welded build-up left HAZ hardness in the 400–490 HV range. With a properly applied temper bead technique, peak HAZ hardness fell to roughly 309–318 HV — comparable to welds that had received full PWHT (met10010099). That is the difference between a joint that will service-harden and crack and one that behaves like a properly treated weldment.

Half-Bead and Two-Layer Techniques

Two classical variants do the work:

  • Half-bead method: the first layer is deposited, then half of it and its HAZ are ground off. The remaining material plus the underlying HAZ is then tempered by the second layer. It is conservative but grinding-intensive.
  • Two-layer (multi-layer) method: no grinding. Bead placement, weaving and heat input are controlled so that each pass falls within the tempering range of the bead beneath it, and no untempered HAZ remains at the finished surface.

ASME Section IX recognizes these approaches through the temper bead welding rules, which convert ordinary Procedure Qualification Record variables into something much stricter. Under AWS D10.8 and Section IX, bead position, bead size, number of layers and heat input become essential variables. Move a bead a few millimetres, change weaving width, or bump amperage outside the qualified range, and the procedure no longer applies. Every variation therefore demands its own mechanical testing — guided bend, tensile and often impact work. This is why temper bead work is expensive to qualify and cheap to ruin.

Controlled Fill for Grade 91

Grade 91 is the case where improvisation is most dangerous, and EPRI’s field-repair guidance is the reference document. EPRI assesses three best-option repairs for components that cannot receive full PWHT:

  1. Matching 9Cr-1Mo-VNbN filler with controlled fill and a low-temperature PWHT (minimum 1250 °F / 675 °C).
  2. 9Cr-1Mo filler with controlled fill and no PWHT.
  3. Nickel-base filler with controlled fill and no PWHT (EPRI repair guidance).

EPRI defines controlled fill precisely: the first layer is deposited with at least 50% bead-to-bead overlap, and subsequent passes follow a predetermined sequence that prevents distortion, promotes tempering of prior layers, and minimizes the risk of reheat cracking. The sequence is written down before welding starts — it is not a welder’s improvisation.

Short Note: In repair work, undermatching soft fillers such as E7018 frequently outperform matching ones. The buttery, low-carbon weld metal stays ductile, tolerates dilution and restraint, and sits against a hardened HAZ as a built-in crack-arrest layer. Overmatching puts strength exactly where the joint does not need it and hardness exactly where it does.

Nickel-Base “Cold” Fillers

When even controlled-fill 9Cr is impractical, nickel alloys (ERNiCr-3 and similar) act as genuinely cold repair fillers: low hardness, no martensite on cooling, and a large coefficient-of-thermal-expansion mismatch that must be managed by careful bead sequencing. They also cost several times more per pound than steel wire, which is worth budgeting for honestly. For comparison on filler handling and hydrogen discipline, our Alloy Steel Welding Electrodes guide covers storage and rebake practice.

When NOT to Attempt Temper Bead

Skip it when the base metal is in an uncontrolled or hardened condition (above roughly 25 Rc), when the joint carries cyclic fatigue or creep-critical service without engineering review, when you cannot verify interpass temperature instrumentally, or when the procedure cannot be qualified with testing. AWS D10.8 and preheat and interpass guidance both assume disciplined heat control; if your crew cannot hold it, no clever bead pattern will save the joint. Tempting as the shortcut looks, an unqualified temper bead weld is simply an uncontrolled hardened weld with extra steps.

Common Mistakes, Crack Diagnosis and Inspection Discipline

For chromium molybdenum steel, the weld rarely fails at the moment of the arc. More often, it’s the small shortcuts made under pressure that come back as cracks days, or even weeks later. With hardenability on the rise the higher up the Cr-Mo ladder you go, discipline must outweigh convenience at every step.

The Mistakes That Cause the Most Damage

Nearly every avoidable Cr-Mo failure can be traced back to one of the following errors:

  • Skipping preheat on ‘thin enough’ material: Wall thickness isn’t a free pass. Even lighter sections of P22 or 4140 can quench to untempered martensite if allowed to cool too rapidly, creating an incredibly brittle heat affected zone.
  • Letting the joint go cold mid-weld: Stopping for a long break without maintaining your minimum interpass temperature is one of the easiest ways to trap diffusible hydrogen. With Cr-Mo, continuity is safety.
  • Overmatching filler or non-low-hydrogen consumables: Swapping your properly classified, low-hydrogen electrodes for general purpose ones drastically increases your hydrogen load. For help with correct filler choices, reference Alloy Steel Welding Electrodes: Selection, Use & Safety before making substitutions.
  • Exceeding interpass or welding on dirty, oily, or hardened base metal: Contaminants introduce hydrogen, and welding over already hardened material creates a brutal double hardening effect with no room for error.
  • PWHT overshoot above Ac1: For Grade 91 specifically, PWHT temperature overshoot above Ac1 creates alpha ferrite and fresh martensite, both detrimental to weld metal and HAZ creep strength and toughness. This is one of the only PWHT mistakes that can permanently downgrade the creep properties of the entire joint.
  • Using ordinary carbon-steel habits on P22: You cannot treat 2.25Cr-1Mo the same as A36. The cooling sensitivity is far higher, and its susceptibility to delayed cracking demands the preheat, low-hydrogen practice, and code driven PWHT that its lower alloyed counterparts would not require.

Quick Note: A common failure pattern is welding 4140 with a plain 70 ksi wire and no preheat and expecting impact-loaded service life — the hard base and soft weld with an untempered brittle HAZ will not survive. For any impact loaded Cr-Mo component, preheat and the correct, hydrogen controlled filler are requirements, not suggestions.

Crack Diagnosis: Reading The Failure

Before you can fix a crack, you must identify what it is telling you. Each location points back to a specific root cause:

  • Underbead cracks: Often hairline and hidden within the HAZ, these are the textbook sign of hydrogen induced cold cracking. They’re most common when preheat was insufficient or consumables were improperly stored.
  • Toe cracks: Originating at the weld toe where stress concentration is highest, these are usually caused by high restraint, excessive interpass temperatures, or rapid cooling.
  • Transverse cracks: Running perpendicular to the weld bead, these can stem from high residual stresses, overmatching filler, or PWHT that was either skipped, or in the case of creep grades, pushed past its allowable window.

Note: Cr-Mo cold cracks are infamous for their delay. Just because a joint passes visual inspection immediately after welding does not mean it’s sound. This is exactly why inspection timing is just as important as the weld itself.

Inspection Discipline To Catch Them

The goal of inspection for hardenable Cr-Mo joints is to catch delayed hydrogen cracks before the system is pressurized. For this reason, the recommended practice is to perform magnetic particle (ASTM E709) or dye penetrant inspection no sooner than 48 hours after welding on hardenable Cr-Mo joints so delayed hydrogen cracks are caught.

Depending on the code, additional methods may be required. Ultrasonic Testing (UT) is invaluable for finding deeper, subsurface underbead cracks, especially on heavier piping. For a go, no-go baseline, a hardness survey is also used as a proxy check, with many codes setting maximum HAZ hardness limits to prevent the formation of overly brittle martensite in the first place.

Finally, hydrogen control must be upheld all the way to inspection. EPRI recommends SMAW electrodes left outside a rod oven or caddie for more than 4 hours be discarded, and that FCAW reels be discarded if welding is interrupted for more than 24 hours. These aren’t overcautious rules, they’re some of the most cost effective insurance you can buy against a delayed crack.

In Cr-Mo welding, perfection isn’t the goal, prevention is. Catch your mistakes on the checklist, not on a hydrotest.

Workshop Checklist, Safety and Final Word on Cr-Mo Welding

Chromium molybdenum steel will not forgive rushed work. The difference between a reliable Cr-Mo weld and a delayed failure comes down to discipline, not talent. With that in mind, use the checklist below, follow the safety measures for high-alloy consumables, and lock in the habits that keep these welds sound for decades.

The Cr-Mo Jobsite Checklist

Work through this in order for every Cr-Mo joint, no matter how small.

  1. Verify the material. Check the material certificate for the exact grade. For 4140, note if it’s annealed, normalized, or hardened. If hardness is above 25 Rc on 4140, stop work until someone with engineering authority signs off.
  2. Confirm the governing code. Identify if the job falls under ASME B31.1, B31.3, Section VIII, or AWS D10.8. Keep in mind that 4140 is not prequalified under AWS D1.1, so an ASME Section IX or AWS B2.1 qualification must govern.
  3. Match filler and check hydrogen. Select filler specific to the grade and service. Check the H-designation on the actual package, as low-hydrogen consumables cannot be substituted with general-purpose stock.
  4. Use a qualified WPS. Do not improvise essential variables. For example, a qualified SWPS under AWS B2.1 can fix essential variables such as base metal, thickness range, filler, and whether PWHT is applied. Changing any of these in the field invalidates that qualification.
  5. Clean thoroughly. Remove oil, paint, scale, rust, and moisture from at least 2 inches back on both sides. Use an approved solvent and allow it to flash off.
  6. Preheat and verify. Set preheat at least 50 °F above your minimum, verify with a contact pyrometer or calibrated temperature sticks, and extend heat at least 3 inches in every direction through full thickness.
  7. Control interpass temperature. Never allow the joint to drop below minimum preheat or exceed the grade’s maximum. Check between passes with instrumentation, not by hand.
  8. Weld without interruption. Complete root, hot pass, fill, and cap in one continuous cycle while the part stays at or above the required temperatures.
  9. Manage post-heat or PWHT. For 4140 or heavy sections, a controlled hydrogen post-heat at 200–300 °C for 1–2 hours can help drive diffusible hydrogen out. For Grade 91, EPRI recommends post-weld heat treatment within 8 hours of completion of welding.
  10. Inspect at the right time. Perform visual inspection immediately, but delay volumetric testing to allow for delayed cracking. Hardness surveys on the HAZ are a reliable check of cooling and tempering.
  11. Document everything. Record heat numbers, filler lot numbers, preheat and interpass readings, PWHT charts, and all inspector sign-offs.

Quick Note: Print this checklist and tape it to the machine. Most Cr-Mo failures aren’t due to a lack of knowledge, but a skipped step on a rushed job.

Safety: What Changes With High-Alloy Consumables

Cr-Mo welding requires extra precautions beyond general shop safety. ANSI Z49.1, Safety in Welding, Cutting and Allied Processes, along with AWS Safety and Health Fact Sheets, should be used as a baseline for every job.

Heat hazards. Joints at 400 °F or higher can sear through leather within seconds. Use heat-resistant gloves, tongs, insulated tools, clearly mark hot zones, and maintain a fire watch for preheated parts.

Electrode storage. Low-hydrogen electrodes must be stored and rebaked exactly per the manufacturer’s instructions. Use a calibrated holding oven and log batches to avoid under or over-baking.

Fumes and ventilation. Chromium-bearing consumables demand proper fume extraction, not a ceiling fan. For confined spaces, use supplied-air respirators, atmospheric testing, and a trained attendant. Pay extra attention to SDS guidance, especially for 9Cr and nickel-base repair fillers. For broader storage and handling practices, refer to our guide on alloy steel welding electrodes: selection, use and safety.

Habits That Prevent Failures

Every step on the checklist builds on the next. Preheat and interpass control slow cooling to allow hydrogen to escape from hard martensite. PWHT tempers that hardened structure and relieves stress. Timed inspection catches cracks that won’t appear at the end of the pass. None of these steps are optional for most Cr-Mo applications.

All of these controls are qualifiable and learnable. With a qualified WPS, an accurate thermometer, clean low-hydrogen consumables, and the patience to finish a joint hot, Cr-Mo welds can consistently pass radiography, hardness surveys, and decades of service.

Final Word

Cr-Mo welding is entirely manageable when heat control and hydrogen discipline are treated as non-negotiable. The alloy won’t warn you of a mistake. It will wait for a stress riser to find a hairline crack. The checklist above is there to make sure that never happens. Know your grade, confirm your code, match your filler, verify your preheat, and never let the joint go cold.

Short Note: The principles of controlled heat input, cleanliness, and procedure adherence don’t stop at Cr-Mo. To see how this same discipline is applied to even higher consequence work, take a look at our aerospace welding: processes, standards and careers guide.

Frequently Asked Questions

Can chromium molybdenum steel be welded without preheat?

Only thin-wall 4130 chromoly (roughly under 3 mm / 0.120 in) in clean, low-restraint joints can typically skip preheat. Every other Cr-Mo grade — 4140, P11, P22, P5, P91 — needs preheat, typically 100-200 °C depending on grade and thickness, to prevent hard martensite and hydrogen cracking.

What filler metal is best for chromium molybdenum steel?

It depends on the required post-weld condition. ER80S-B2/E8018-B2 suits 1.25Cr-0.5Mo service, E9018-B3/ER90S-B3 suits 2.25Cr-1Mo (P22), B8/B9 classifications suit 9Cr grades, and ER70S-2 or ER80S-D2 is common for as-welded 4130 structures. Matching 4130 filler is only appropriate if the whole weldment will be heat treated afterward.

Does all chromium molybdenum steel require post-weld heat treatment?

Almost all construction codes require PWHT for Cr-Mo pressure components, with limited exemptions for very thin-wall pipe. Small-bore thin-wall 4130 fabrication is often done as-welded. Whether PWHT is mandatory is set by the governing code, specification or engineer, not by the welder.

What preheat temperature does 4140 need?

Commonly 200-300 °C (400-575 °F), rising with thickness and joint restraint, with interpass held between the preheat and roughly 350 °C (650 °F). Never let interpass exceed the base metal’s original tempering temperature.

Why do Cr-Mo welds crack after welding is finished?

Delayed underbead cracking is hydrogen-induced: diffusible hydrogen from moisture, coatings or shielding gas migrates into a hard HAZ under residual stress. It can appear hours to days later, which is why PWHT or hydrogen post-heat is timed promptly and inspection is not done immediately.

What is temper bead welding and when is it used?

Temper bead welding places beads in a controlled sequence so each pass tempers the heat-affected zone beneath it, mimicking PWHT when a furnace or stress relief is impractical. It requires strict procedure qualification because bead size, position and heat input become essential variables.

What is special about welding Grade 91 (P91) steel?

Grade 91 needs 200 °C preheat, low interpass (max ~300-315 °C) so each layer transforms to martensite before the next pass, and PWHT below Ac1 — typically 730-775 °C for at least 2 hours. Its failure to transform properly leads to high hardness, Type IV HAZ cracking and premature creep failure.

Why is the IIW carbon equivalent formula not used for Cr-Mo steels?

Because the standard IIW CE model was calibrated for carbon and structural steels; chromium and molybdenum contribute to hardenability in ways it underestimates at these alloy levels. Cr-Mo procedures rely on grade-specific preheat tables and code requirements instead.

NH

Nicholas Harper

Editorial Team

Covering welding processes, equipment selection, and shop-floor practice for WeldZen — with every claim checked against the governing code.

Share this technical standard

Related Studies

Technical Dispatch

Get The Next Benchmark First

New field studies and calibration sheets, once a week. No spam.