Welding Ductile Cast Iron: The Complete Repair Guide

A machinist once told me he spent eleven hours building up a worn pump housing boss, only to watch a hairline crack open across the repair as the part cooled on the bench overnight. The weld looked perfect when he struck the arc off. Ductile iron forgives nothing about how you heat it, how you cool it, or which rod you grabbed off the rack — and understanding why is the difference between a salvageable casting and scrap.

Why Welding Ductile Cast Iron Is a Different Problem

A rebuild shop in Ohio took in a split valve body from a municipal water pump. The casting was twenty years old, discontinued by the OEM, and a replacement would have cost four figures plus eight weeks of downtime. The shop had one chance to put it back in service, and everyone in the room knew that a bad weld on cast iron does not fail politely — it fails quietly, hours after the arc is struck, when the part is already bolted back together or, worse, already machined to size. That quiet, delayed failure is what makes ductile iron welding feel like a different trade altogether.

So let’s answer the principal question up front, because you should not have to read two thousand words to get it: yes, ductile iron is weldable, and it is actually more weldable than gray cast iron — provided you respect its metallurgy. The casting will not tolerate the casual approach that works on mild steel, but it rewards a disciplined one. The same nodular graphite structure that gives ductile iron its name also makes it the more forgiving of the cast irons to repair, because the graphite dissolves less carbon into the weld pool during the thermal cycle, which means fewer carbides and less high-carbon martensite at the joint, according to the Ductile Iron Pipe Research Association field welding guidelines.

What People Actually Want When They Search for Welding Ductile Cast Iron

In practice, this is almost never a fabrication topic. As the AWS D11.2 Guide for Welding Iron Castings makes clear, welding cast iron is predominantly a repair discipline. The three jobs that fill the schedule are defect salvage (blowholes, cold shuts and inclusions found after machining), worn-surface build-up (pump sleeves, bearing bores, shaft journals, gear teeth), and machining-error correction (an undersized bore or a mangled bolt pattern). A fourth, less common job is joining ductile iron to steel, which behaves like two different materials having an argument under restraint.

Whatever the job, three things go wrong, and nearly every cast iron failure traces back to one of them:

• Carbon pick-up. The arc dissolves graphite from the base metal into the weld pool. The re-solidified zone ends up enriched in carbon and, on cooling, forms hard iron carbides instead of graphite — a layer of white iron right at the fusion line.
• HAZ martensite. Carbon that migrates out of the heat-affected zone converts that zone into high-carbon martensite and bainite, which is hard, brittle and unforgiving of restraint.
• Shrinkage stress. As the weld and the surrounding metal contract, a restrained casting stores tensile stress. If that stress exceeds the strength of the brittle zones before they can yield, you get a crack — often hours later.

Where You Will Run Into It

Ductile iron shows up wherever a part needs strength, damping and castability at a sane price:

• Pump and valve bodies, bonnets, covers and impellers in water and wastewater service
• Heavy equipment bushings, bosses, frames and linkage castings
• Ductile iron pipe fittings and specials, both in the plant and in the field
• Engine blocks, cylinder heads, gear cases and flywheels
• Dies, jigs, fixtures and tooling where a scrapped casting costs real production time

Each of those applications brings its own restraint, section thickness and service stress, and as you will see, those variables matter as much as the alloy itself. The rest of this guide works through the material science first — what the graphite nodules actually do during the thermal cycle — then moves into filler metal selection, preheat and post-weld heat treatment, process choice, a step-by-step shop procedure, troubleshooting, the governing standards, and finally the honest answer to when a casting should not be welded at all. If you are new to stick welding as a repair process, our guide to why SMAW still wins covers the fundamentals before you get to the nickel rods.

Keep one sentence in mind through all of it: ductile iron forgives nothing about how you heat it, how you cool it, or which electrode you pulled off the rack — but a welder who understands why will reliably succeed where others scrap the part.

Ductile Iron Basics: What the Nodules Do During Welding

Before you strike an arc on a ductile iron casting, you need to know what you are actually heating. Ductile iron — also called nodular iron, spheroidal graphite (SG) iron, or ductile cast iron — is not one material but a family of closely related alloys, and the differences between them decide whether your repair holds or cracks. The same graphite structure that gives the metal its name is also why it welds better than gray iron. The catch is that melting destroys that structure, and it never returns on its own.

From Flakes to Spheroids: Why Morphology Rules Weldability

Most ductile iron starts from a chemistry that looks almost identical to gray iron: roughly 3.5% carbon and 2.4% silicon by weight, with iron making up the balance along with small additions of manganese, sulfur, and residual elements. Carbon at that concentration cannot stay fully dissolved in solid iron, so it precipitates as graphite during solidification. The shape of those precipitates is the whole ballgame. In gray iron the graphite forms sharp-edged flakes that act like internal cracks; in ductile iron a small magnesium addition (sometimes cerium or another spheroidizer) changes the surface tension so the graphite grows into smooth spheres. The nodules blunt stress concentrations and let the surrounding metal deform, which is why ductile iron can reach elongations in the 10–20% range while gray iron essentially has none.

Nodule count and the matrix around it define the common grades:

Grade family	Matrix	Character	Welding behavior
Ferritic ductile iron	Ferrite (soft, low-carbon)	Best ductility and thermal conductivity	Most forgiving to weld; least hardening in the HAZ
Pearlitic ductile iron	Pearlite (ferrite + iron carbide)	Higher strength and hardness	More martensite, higher crack susceptibility
Ferritic–pearlitic	Mixed	General-purpose engineering grades	Moderate; intermediate response
Austempered ductile iron (ADI)	Ausferrite	Excellent strength-to-weight, toughness, fatigue resistance	Weldability is a limiting factor in manufacturing

Ferritic vs Pearlitic: Which One Cracks More Readily

The matrix matters more than most repair shops realize. In a ferritic grade, most carbon already sits harmlessly in graphite nodules, so relatively little is available to harden when the heat-affected zone (HAZ) cools quickly. In a pearlitic grade, a meaningful share of that carbon is locked up as iron carbide within the pearlite. When the arc passes and the zone behind it quenches against cold surrounding metal, that carbon transforms into high-carbon martensite and bainite. As Modern Casting’s overview of gray and ductile iron welding puts it, pearlitic ductile iron produces more martensite in the HAZ than ferritic grades and is significantly more crack susceptible during welding. The same principle governs hardening in carbon and alloy steels, which is why electrode guides such as our Alloy Steel Welding Electrodes: Selection, Use & Safety stress hydrogen control and cooling rate — the variables you will also be managing here.

Austempered ductile iron deserves a specific warning. ADI’s combination of strength, toughness, and fatigue performance is outstanding, and it increasingly replaces steel in demanding parts — but welding heat destroys the carefully controlled ausferritic structure, and the rapid thermal cycle produces a hard, brittle zone that is difficult to avoid. Research on ADI weldability consistently identifies this as a limiting factor in manufacturing, so many ADI components are better designed out of a weld repair than into one.

The Three Zones Every Ductile Iron Weld Creates

Whatever process you choose, an arc on ductile iron creates three distinct regions, and each responds differently:

• Fusion zone — the metal that actually melted and mixed with the filler. This is where carbon pick-up from the base metal is highest.
• Heat-affected zone (HAZ) — base metal that never melted but got hot enough to transform. Hardening, residual stress, and most cracking originate here.
• Unaffected base metal — original casting beyond the thermal reach of the arc, still carrying its nodule structure and properties.

The critical point is what happens inside the fusion zone. Once ductile iron melts, the magnesium that spheroidized the graphite is partly burned off or oxidized in the arc, and the cooling rate is far too fast for nodules to re-establish. As documented in Vulcan Hammer’s VFE RWP-3 ductile iron repair welding procedure, graphite in the re-solidified weld metal re-precipitates in vermicular (worm-like) or quasi-nodular forms rather than true spheres, cutting ductility and impact resistance by roughly 50%. In other words, the deposited weld metal is never again ductile iron — it is a weaker, coarser cousin, with a hard white-iron band possible right at the fusion line. Knowing that shapes every decision that follows: keep dilution low, control heat input, and build the joint so the ductile properties you need live in the filler metal and the unaffected base metal, not in the melted puddle itself.

That is the metallurgy in a nutshell. The sections ahead turn it into procedure: filler selection, preheat, technique, and the heat treatment that lets these three zones settle down instead of tearing apart.

The Metallurgy of Failure: Why Ductile Iron Cracks

Every failed ductile iron weld follows the same three-act script: carbon leaves the nodules near the fusion zone, the heat-affected zone hardens into something brittle, and cooling tries to pull the joint apart. Understand where each mechanism strikes and you can design a repair that survives. Ignore one and you get the result from our introduction — a good-looking bead with a crack through it by morning.

Mechanism One: Carbon Walks Out of the Nodules

Graphite nodules are essentially carbon reservoirs sitting in a steel matrix. When the arc comes within a few millimetres, the metal at the fusion line sees 1,200–1,300 °C, and at that temperature graphite dissolves into austenite instead of staying as discrete nodules. The dissolved carbon then diffuses down a steep gradient into the cooler base metal and, to a lesser extent, into the weld pool itself.

The first consequence is a thin band at the edge of the fusion zone that has picked up so much carbon it re-solidifies as iron carbide — cementite, with essentially no graphite left in it. That band is cast iron’s white iron: extremely hard, glassy, and brittle. The (DIPRA field welding guidelines) describe exactly this sequence: graphitic carbon dissolves and diffuses into iron, so carbides form at the fusion zone edge and high-carbon martensite and bainite form through the rest of the heat-affected zone — which is what drives crack susceptibility in ductile iron.

The second consequence lands in the weld pool: the filler is diluted with carbon it was not designed to carry. This is where alloy steel welding electrodes earn their bad reputation on cast iron. Steel fillers containing chromium and manganese promote hard, brittle carbides, because incoming carbon meets chemistry that stabilizes carbides rather than rejecting graphite. A 2022 review of filler materials for ductile cast iron found iron-based fillers with Cr and Mn unsuitable for this reason (MATEC Web of Conferences, 2022).

Mechanism Two: Martensite and Bainite in the HAZ

Away from the fusion line the metal never melts, but it still reaches austenitizing temperature. Because that zone is carbon-enriched by diffusion, it has lost the low-hardenability character of ordinary iron, so when the arc moves on and the casting acts as a heat sink, austenite transforms to martensite and bainite instead of ferrite.

What you get is a hardness gradient. A typical ferritic ductile iron base sits around 150–250 HV; the coarse HAZ on a fast-cooled weld made with a steel filler can spike well above 500 HV. Martensite at that hardness has almost no ductility, and the transformation involves a volume increase that adds tensile stress inside metal already restrained by the casting.

Ferritic grades are the most forgiving, since they start with less carbon in solution. Pearlitic grades harden faster and deeper, and austempered ductile iron (ADI) is worse still: the weld cycle re-austenitizes a carefully engineered microstructure and replaces it with untempered martensite. Keep that in mind when someone hands you an ADI gear and asks for a quick build-up.

Mechanism Three: Shrinkage Stress Against a Metal That Won’t Yield

Every weld deposit contracts as it cools, both from solidification shrinkage and thermal contraction. In a stiff casting that contraction is resisted, and resistance means stress. The critical difference from mild steel is at the yield point: cast irons lack ductility, so once loaded beyond yield they fracture rather than deform. Steel bleeds off local stress by yielding; ductile iron stores elastic strain until the weakest link gives way — almost always the brittle white iron band at the fusion line.

This is why restraint matters as much as chemistry. A long bead in a heavy, rigid section accumulates far more stress than a short bead in a thin flange, even with identical filler and amperage. It is also why weaving is discouraged: wide, shallow stringer beads flex less as they cool.

Hot Cracking vs. Cold Cracking: And Why Cracks Show Up at 2 a.m.

Hot cracking happens during solidification or within minutes of it. The same MATEC review identified hot cracking as the most serious issue with nickel-alloy fillers, usually right at the fusion line as summarized in the filler comparison. Wide beads, excessive heat input, and low-melting segregated films are the usual contributors.

Cold cracking, also called delayed or hydrogen-induced cracking, is the one that surprises people. It needs three things at once: hydrogen, a hard microstructure, and tensile stress. Hydrogen comes from damp electrode coatings, oil-soaked castings, rust, and shielding gas impurities. It diffuses freely through hot weld metal, then concentrates at trap sites — graphite nodule interfaces, notch tips, and bead toes. Once enough collects where martensite and residual stress are waiting, the crack opens. Hours later. Often overnight. Usually along the fusion line, at a bead toe, through a crater, or across an abrupt change in section.

The practical takeaway: a weld that looked flawless when the arc was struck may still be chemistry in progress. Controlling hydrogen is not optional — dry electrodes, degreased castings, and slow cooling, all covered in the heat treatment and procedure sections ahead. Managing these mechanisms comes down to filler choice, controlled heat, and joint design, and the filler is where the battle is won.

Choosing the Right Filler Metal for Ductile Iron

Every successful ductile iron repair starts at the rod rack, not the arc. Filler choice drives how much carbide forms at the fusion line, how much the weld shrinks on cooling, and whether the finished bead machines cleanly or eats a carbide cutter in twenty minutes. Cast iron filler metallurgy is a solved problem: AWS A5.15 classifies a family of cast iron electrodes built for exactly this job. Once you understand what the nickel percentages actually do, the choice becomes almost automatic.

Why Nickel Is the Magic Element

Steel weld metal is saturated with iron, and iron loves carbon. When a steel bead sits on ductile iron, carbon migrates from the base metal into the puddle, the puddle rejects it as iron carbide, and you get a hard white cast iron layer right where cracks start. Nickel changes that chemistry. Nickel is a graphitizer: it stabilizes graphite rather than carbide, so carbon dissolving into a nickel-rich puddle tends to come back out as graphite instead of cementite. That single fact explains the whole classification system. AWS A5.15 recognizes three main groups:

• ENi-CI (about 93% nickel) – the high-nickel “pure nickel” rod.
• ENiFe-CI (44–55% nickel, balance iron) – the nickel-iron workhorse.
• ENiCu-A / ENiCu-B – copper-nickel rods for specific applications.

ENiFe-CI: The Default Answer for Ductile Iron

If you take one thing from this section, make it this: 55% nickel-iron (ENiFe-CI) is the default filler for ductile iron repairs. The reason is strength. Deposited ENiFe-CI metal runs about 50,000 psi (345 MPa) tensile strength, close enough to ductile iron grades like 65-45-12 and 80-55-06 that the weld is not the weak link in the casting.

It also runs at lower current and puts less heat into the part than steel electrodes, which means less carbide and martensite in the heat-affected zone, as documented in VFE’s RWP-3 ductile iron repair procedure. Typical settings for a 3/32 in (2.4 mm) electrode are 60–90 amps DCEN; a 1/8 in (3.2 mm) rod runs roughly 90–110 amps. Keep the arc short and the beads small.

ENi-CI: The Machinability Specialist

The 93% nickel rod is softer, more machinable, and gentler on cutting tools, with a lower expansion coefficient closer to cast iron itself. The trade-off is tensile strength in the range of roughly 25,000–35,000 psi (170–240 MPa) and a much higher price per pound. Use ENi-CI when the finished surface will be machined, when distortion must be minimized on a delicate casting, or as a final cosmetic pass over an ENiFe-CI root. For structural build-up on a loaded boss or bushing, it is the wrong tool.

Filler Selection at a Glance

Filler	Ni Content	Approx. Strength	Best Used For
ENiFe-CI (55% Ni)	44–55%	~50 ksi (345 MPa)	Default for ductile iron; structural build-up, groove welds, machinable surfaces
ENi-CI (93% Ni)	~93%	~25–35 ksi (170–240 MPa)	Machinability, thin/delicate castings, final passes, gray iron
ERNiFeMn-CI (MIG)	~47%	Comparable to ENiFe-CI	Automatic/GMAW build-up and continuous deposits
Alloy 625 (ERNiCrMo-3)	~60%	Very high	GTAW root pass and buttering layers on critical joints
E7018 steel rod	0%	~70 ksi (480 MPa)	Not recommended for ductile iron; cracking risk at interface
Bronze braze rod	Variable (Ni-bronze)	Brazed joint	Braze-welding where base metal must not melt

Specialty Options: ERNiFeMn-CI and Inconel 625

For MIG and flux-cored build-ups, ERNiFeMn-CI is the dedicated cast iron wire — roughly 47% nickel with manganese additions that help manage hot shortness. It is designed for continuous deposition on large worn surfaces where stopping to change rods every three inches would take all day. Flux-cored cast iron wires exist too; if you are new to the process, our aluminum flux core welding guide explains the short-circuit transfer fundamentals that apply across wire types.

Alloy 625 (ERNiCrMo-3) shows up in research and pressure-critical work as a root or buttering filler. A ductile iron weldability study in Metals used a GTAW root with Inconel 625 followed by 97.6% nickel hot passes, producing acceptable joint strength; a subsequent 900°C anneal dissolved the carbides and restored a ferritic matrix with nodular graphite. Buttering the cast faces with a nickel filler first, then welding the structural pass to the buttering, is a legitimate technique for high-value castings.

Why Steel Electrodes Fail on Ductile Iron

E7018 is cheap, familiar, and in this application, a gamble. Without nickel to stabilize graphite, carbon pickup at the fusion line forms a hard carbide band, and the higher heat input typical of steel electrodes creates more high-carbon martensite in the HAZ. DIPRA’s guidelines state plainly that mild-steel arc electrodes are not recommended for ductile iron pipe because of the high risk of cracking at the weld-to-base-metal interface — and the same logic applies to any ductile iron casting. Steel rod on ductile iron is the single most common cause of the overnight cracked repair. For a broader comparison of rod families, see our guide to alloy steel welding electrodes.

Dilution, Amperage and Practical Rules

Dilution — how much base metal melts into your bead — is the dial you are really tuning. Low current, a short arc, stringer beads, and nickel buttering layers all push dilution down, and lower dilution means fewer carbides at the fusion line. Run DCEN for most nickel cast iron rods, stay near the bottom of the manufacturer’s amperage range, and never weave wider than about two electrode diameters. Lincoln Electric’s cast iron guidelines reinforce the same recipe: minimal penetration, minimal dilution, minimal heat.

Choose ENiFe-CI for the job, reserve ENi-CI for finishing, keep steel rod off the casting, and half your cracking problems never get a chance to start. Next comes the heat that surrounds all of this: preheat, interpass control, and post-weld treatment.

Preheat, Interpass Temperature and Post-Weld Heat Treatment

If filler metal selection is the first decision in a ductile iron repair, heat management is the one that decides whether the job survives the night. Almost every failed cast iron repair story ends the same way: the weld looked fine, the part was allowed to cool on an open bench, and a crack showed up hours later. Preheat, interpass control and post-weld heat treatment are how you break that pattern.

Deciding Whether You Need Preheat At All

Here is the part that surprises people: most ductile iron castings do not require preheat. A small, lightly restrained repair — a worn boss, a shallow groove, a non-critical build-up with plenty of machining allowance — can often be welded cold with a nickel-iron electrode and slow cooling under a blanket. Preheat is reserved for large castings, thick sections, deep grooves, joints with high restraint, or parts that cracked once already and are being given a second life.

When preheat is required, published guidance puts typical minimum preheat temperatures for cast iron at roughly 200–750°F (93–398°C), depending on section thickness, grade and restraint. In practice, most repair procedures fall into three bands:

Situation	Typical preheat
Small, unrestrained, non-critical build-up	Cold to about 150°C (300°F)
Moderate repair, machined groove, light restraint	200–315°C (400–600°F)
Heavy, complex or highly restrained casting	Up to about 500°C (930°F), applied uniformly

Preheats up to 315°C are the standard recommendation for large or highly restrained components; below that threshold, you are usually paying for heat you did not need.

Furnace Preheat vs Localized Preheat

A furnace preheat is the gold standard because the whole part reaches temperature evenly, which removes the differential expansion that starts cracks before the arc ever strikes. A published ductile iron repair procedure from Vulcan is a good model to copy: start the furnace at 100°F (40°C), rise at 100°F (about 40°C) per hour, reach 550°F (290°C), then hold one hour per inch of section thickness, with a six-hour maximum hold.

Localized preheat with an oxyfuel torch or ceramic pads is acceptable only when the whole part cannot fit the furnace. Heat a wide band well beyond the weld zone, work slowly and evenly, and monitor with a temperature stick or infrared thermometer rather than by color. Uneven heating is worse than no heating.

Interpass Temperature: The Ceiling That Gets Ignored

Preheat sets the starting temperature; interpass control decides whether you keep it. In the Vulcan procedure the maximum interpass temperature is 700°F (370°C), with 600°F (320°C) preferred. Treat 370°C as a hard ceiling, not a target: let the part fall below it between passes, clean the slag, and keep the beads short and stacked like stringer beads rather than long, hot, weaving runs. If you are running stick, our Advantages of Stick Welding guide explains why SMAW’s low, controllable heat input suits this kind of work so well.

Post-Weld Heat Treatment and Slow Cooling

Post-weld heat treatment is recommended for ductile iron castings rather than treated as optional. AWS D11.2/D11.2M sets out the general framework, and the practical split is by grade: ferritic grades are stress relieved near 600°C, while pearlitic grades are usually normalized and tempered to restore a uniform structure. A simple stress relief hold in the 540–600°C range, followed by furnace cooling, handles most repair work.

Cooling is where good jobs die. Follow every weld with an insulating blanket, dry sand bed or controlled furnace cool so the part drops temperature slowly and evenly. Lincoln Electric’s cast iron welding guidelines are blunt about it: no air blasts, no water quenching, no cold drafts. Air-cooling a ductile iron repair can harden the heat-affected zone into martensite faster than most welders expect, and the crack will arrive quietly, hours after the part has been put away.

Note: Plan the cooling path before you strike the arc, not after. Have the blanket, sand box or furnace ready and at temperature, because the repair window between “finished welding” and “too late to insulate” is shorter than it feels in the shop.

Process Selection: SMAW, GMAW, GTAW, FCAW and Braze-Welding

Every arc process can put metal on a ductile iron casting. That is not the real question. What separates a successful repair from a cracked one is how much base metal you melt, how quickly heat travels in and out of the joint, and how much control you have over each individual bead.

SMAW: The Industry Default for Ductile Iron Repair

Shielded metal arc welding with nickel-iron electrodes is the most common technique for welding ductile iron, and it has held that position for decades. The reasons are practical: the equipment is cheap and portable, a single electrode class works in flat, horizontal, vertical and overhead positions, the flux coating shields the dirty, oxidized surfaces you inevitably find on old castings, and the short arc keeps heat localized. You can read more about the fundamentals in our guide to the advantages of stick welding.

The settings matter as much as the rod. Nickel-iron electrodes such as ENiFe-CI are best run on DC reverse polarity with a short arc length. The DIPRA field welding guidelines go further and recommend that weave width should not exceed three electrode core-wire diameters, which keeps the molten pool narrow and the heat spread into the casting to a minimum (DIPRA). Use the smallest diameter electrode that will carry the current you need, because a thinner bead means a smaller heat-affected zone.

GTAW: Maximum Control, but No Room for Sloppy Prep

Gas tungsten arc welding gives you something SMAW cannot: precise, foot-pedal-controlled heat input and the ability to add filler a drop at a time. Every gram of filler you do not add is base metal you do not have to dilute, which is exactly the game you are playing with ductile iron. GTAW with ERNiFe-CI or a high-nickel filler produces clean, slag-free roots that need minimal machining and is the process of choice for thin sections, small intricate repairs, and any job where the finished weld has to be machined. The trade-offs are cost, speed, and skill. There is no flux to buffer an unsteady hand, shielding gas coverage is unforgiving outdoors, and the process is slow enough that it rarely makes sense for large build-ups.

GMAW and FCAW: Low Heat Input, Real Fusion Risk

Short-circuit transfer GMAW genuinely minimizes heat input, which sounds ideal until you remember what casting surfaces look like. Graphite nodules, sand inclusions, oxide film and micro-porosity all conspire against fusion, and the short-circuit arc tends to sit on top of the surface rather than biting into it. Lack of fusion is the signature defect of MIG on cast iron. Flux-cored wires bring back the flux advantage and the portability of stick, which has made self-shielded FCAW popular on heavy equipment in the field. If you go this route, keep the wire diameter small, use the lowest usable current, and consider a buttering layer of nickel-iron first so the MIG bead fuses to a clean surface instead of raw casting.

Oxyfuel Fusion Welding and Braze-Welding

Oxy-acetylene fusion welding of cast iron generally requires preheating to a dull red heat around 650°C with a neutral or slightly reducing flame, which gives you a large, uniform, slow-cooling puddle (Cast Iron Welding Procedures). The flame chemistry matters: an oxidizing flame burns carbon out of the puddle and pushes you toward white iron.

Braze-welding with bronze or nickel-bronze filler is the genuinely low-stress option because the base metal never melts. You get a metallurgical bond through wetting rather than fusion, minimal dilution, and almost no risk of HAZ martensite. The honest limits are that bronze is weaker and softer than the parent iron, it has an upper service temperature ceiling, and the repair is a different color from the casting.

The Settings Philosophy: Low Heat, Short Beads, Everywhere

Whatever process you pick, the discipline is the same: small diameter consumables, low current, short arc, stringer beads rather than wide weaves, skip-weld from opposite ends toward the middle, clean and peen every pass, and respect the interpass ceiling. The Daiko NiFe-CI guidance is explicit that for spheroidal graphite grades, a buttering procedure using low heat input and low-temperature technique is advised to avoid hot cracking in the heat-affected zone.

Quick Process Comparison

Process	Best for	Main risk
SMAW (ENiFe-CI)	Most repairs, field work, all positions	Slag inclusion if passes are not cleaned
GTAW	Thin sections, machined surfaces, buttering	Over-reliance on welder skill, slow
GMAW short-circuit	Small build-ups with good access	Lack of fusion on cast surfaces
FCAW self-shielded	Heavy equipment, outdoor repair	Porosity, high heat if mis-set
Oxyfuel fusion	Large castings needing uniform preheat	Oxidizing flame burns carbon out
Braze-welding	Stressed or thin castings, machinability	Lower strength, temperature limits

A Step-by-Step Ductile Iron Repair Procedure You Can Follow

Every rule in this article only pays off when it happens in the right order on the shop floor. This is the sequence experienced repair shops actually run, from the first look at the casting to the final inspection report. Work through it in order and resist skipping ahead: each step protects the next one.

Steps 1–3: Identify, Clean and Degas the Casting

1. Identify the grade and the service stresses. Look for foundry marks, grade stamps such as 65-45-12 or 80-55-06, and any “do not weld” or “repair prohibited” marking. Ask what the part does: a pump casing in mildly restrained, low-fatigue service is a good candidate; a safety-critical component in fatigue duty may not be. If you cannot identify the alloy or the part history, find out before you strike an arc.
2. Clean and degrease thoroughly. Strip grease, paint, scale and process residues with a solvent. Contaminants are the fastest route to porosity and weld-metal contamination, and they are far easier to remove cold than to deal with as blowholes later.
3. Degas oil-saturated castings. Castings that have spent years in service can hold oil and hydrocarbons deep in the graphite network; heating drives them out as gas and produces porosity in the deposit. Cast iron repair training material recommends a uniform degassing cycle at about 370°C (700°F) for roughly 30 minutes on oil-saturated parts, then cooling or going straight into preheat.

Steps 4–6: Excavate the Defect and Prepare the Groove

4. Remove the defect completely. Chip, grind or machine out the cracked or porous metal until you reach sound material. Air arc gouging and carbon arc techniques speed excavation on heavy sections, but follow with grinding to cut the affected layer and any white iron left by the arc.
5. Inspect the clean cavity. A dye-penetrant or magnetic-particle check before welding confirms that the crack you can see is the whole crack, and that the walls are sound metal rather than a hidden tail running out of sight.
6. Cut a V or U groove and remove the casting skin. Machine or grind a groove generous enough for proper fusion, and cut through the skin. As Modern Casting explains, castings carry an oxide-rich surface layer that resists melting and mixing with weld metal, so leaving it in place guarantees lack of fusion.

Steps 7–9: Preheat, Butter and Lay the Beads

7. Preheat uniformly. Bring the whole part, not just the repair area, up to the preheat your procedure calls for and hold it. Localized hot spots create exactly the thermal gradients that generate cracking stress.
8. Butter the joint faces first. Lay a thin, high-nickel layer on the groove faces before filling. Buttering buffers dilution between cast iron and your filler, so the first structural bead welds onto nickel-rich metal instead of raw cast iron.
9. Run short stringer beads, alternating positions. Deposit roughly 50 mm (2 in) of bead at a time, then move elsewhere and let that area cool slightly, building wide cavities from the sides inward. Keep arc length short and current low, and avoid wide weaves that put a large molten pool into a cold, restrained casting. SMAW’s advantages suit this work well, provided your electrodes are fresh and stored dry as low-hydrogen practice expects.

Steps 10–13: Peen, Cool, Machine and Inspect

10. Clean and peen every pass. Chip and wire-brush all slag before the next bead, and peen the hot bead before it cools too far. Peening stretches the shrinking surface to offset contraction stress, but stop once the metal hardens: hammering cold weld metal only risks damage.
11. Control interpass temperature. Keep a thermometer or temperature stick on the part and let each area drop back into the acceptable interpass window before re-striking nearby. Do not chase speed.
12. Slow cool with insulation. Bury the finished part in dry sand, cover it with a ceramic fibre blanket, or return it to the furnace. Air cooling, or worse a water quench, will harden the heat-affected zone and hand you a crack hours later.
13. Optional stress relief, then machine and inspect. A post-weld stress-relief cycle is worthwhile on heavy or complex castings. Machine back to drawing dimensions only after cooling is complete, then run a final dye-penetrant or magnetic-particle inspection before the part returns to service.

Follow this sequence once carefully and it becomes muscle memory. The casting tells you when you have rushed it, usually somewhere between the third bead and the next morning.

Common Ductile Iron Welding Problems, Mistakes and Fixes

Most ductile iron repairs fail for the same few reasons. Instead of re-running a hotter pass, it’s far more effective to fix the process. Here are the issues most likely to appear and how to correct them.

1. Cracks Hours After Welding

Cause: Delayed hydrogen cracking. Damp electrodes, heavy restraint, and no preheat leave a hard, brittle heat-affected zone, with cracks typically starting at the weld toe.

Fix: Bake low-hydrogen rods, preheat uniformly to 200–315°C (400–600°F), hold a steady interpass temperature, run short stringer beads in alternating positions, and cool the part under an insulating blanket. For this type of repair, nickel-iron SMAW is often the most forgiving route. Hold the casting at least overnight before inspection. Guidance for welding and inspecting iron castings can be found in AWS D11.2/D11.2M, Guide for Welding Iron Castings.

2. Hard White Layer and Unmachinable Beads

Cause: Excessive heat input and fusion. Carbon from the base metal can form iron carbide (white iron) along the fusion line, creating an unmachinable surface.

Fix: Switch to a 55% nickel ENiFe-CI filler, use smaller diameter electrodes at lower current, and keep a short arc. It’s important to choose the electrode deliberately rather than grabbing a general purpose rod. For added protection, butter the joint faces with a soft nickel-iron layer to fuse into deposited metal instead of raw casting.

3. Porosity and Blowholes

Cause: Contamination. Oil-soaked castings, damp coatings, dirty electrodes, improper shielding gas, or trapped slag can all introduce gas pockets into the weld.

Fix: Degrease and dry the casting (bake out any oil saturation), store electrodes correctly, verify gas flow, and chip and wire-brush every pass until bright before depositing the next bead.

4. Cracking From Excessive Base Metal Fusion

Cause: Dilution. Long, wide weaves and high current dissolve too much base metal into the pool, raising its carbon content and causing it to crack on contraction.

Fix: Lower heat input, shorten the arc, travel faster, and lay stacked stringer beads no wider than roughly three times the electrode diameter. Alternate positions to avoid hot spots. If a bead must be removed, grind it as air arc welding and carbon arc gouging are rarely appropriate for ductile iron removal.

5. Peening Damage

Cause: Over-peening. While light peening can offset shrinkage, heavy or late blows can hide cracks and work-harden the surface. As noted in DIPRA’s field welding guidelines, force must be enough to deform the deposit without rupturing it.

Fix: Only peen while the bead is hot, use light, uniform blows across its full length, stop as soon as it’s slightly spread and smooth, and never peen a cold, finished pass.

6. Welding in Highly Stressed Areas

Cause: Restraint. According to the VFE RWP-3 ductile iron repair procedure, ductile iron should never be welded in highly stressed parts or areas, as contraction will drive cracking with no filler able to compensate for it.

Fix: Relocate the repair if possible, stress-relieve as appropriate, or opt for a non-fusion method such as bronze braze-welding or mechanical sleeving.

7. Excessive Weld Volume

Cause: Overbuilding. VFE RWP-3 limits deposited weld volume to below 20% of the metal thickness. Beyond that, shrinkage stress and HAZ damage outweigh any benefit.

Fix: Stop building. Consider machining and bushing, mechanical sleeving, nickel-bronze braze-welding for buildup, or full replacement if the repair would require multiple times its thickness in weld metal.

8. Rapid Cooling

Cause: Quenching. Blowing air or water onto a hot weld cools it far too quickly, creating a brittle joint.

Fix: Prioritize slow cooling. Bury the part in dry sand, wrap it in a ceramic insulating blanket, or allow a controlled cooldown. As Modern Casting states, slow cooling is the rule, a point that matches Lincoln Electric’s cast iron guidelines.

Standards, Codes and Qualification: AWS D11.2, A5.15, ASME and Ductile Iron Pipe

Plenty of ductile iron repairs never see a specification. A worn bushing boss gets built up, machined, and back into service the same week. But the moment a casting is pressure-retaining, safety-critical, or owned by a customer with a quality system, someone will ask which procedure you followed, who qualified it, and where the record is. This section is the map of those documents.

The Documents That Govern a Cast Iron Weld Repair

Start here. These are the standards you will actually cite in a written procedure or an inspection report:

Document	What it tells you
AWS D11.2/D11.2M:2025, Guide for Welding Iron Castings	The core guide for welding gray, ductile and malleable iron: joint design, preheat, filler choice, PWHT and inspection. Approved by ANSI on September 17, 2025, it revises the old AWS D11.2-89 (R2006), so any shop copy from before then is out of date.
AWS A5.15	Specification for welding electrodes and rods for cast iron — the classification and testing basis for the ENiFe-CI and ENi-CI consumables covered earlier in this guide. Our related alloy steel welding electrodes guide explains how classification specs work in practice.
AWS B2.1	Standard for specification and qualification of welding procedures and performance qualification — the rules for how a WPS gets written and how a welder gets tested.
ASME Boiler and Pressure Vessel Code, Section IX	Governs welding qualification for pressure work and supplies form QW-482, a suggested format for welding procedure specifications.
MIL-STD-248	Welding, brazing and bonding classifications used across defense and aerospace procurement.
DIPRA field welding guidelines	Field welding and cutting of ductile iron pipe.
ANSI Z49.1 / OSHA	Welding, cutting and allied processes — safety, ventilation, PPE and fire control.

PQR, WPS and WQT: The Three-Legged Stool

A WPS (Welding Procedure Specification) is the written recipe: base metal group, filler classification, current and polarity, preheat and interpass limits, heat input, joint design and post-weld treatment. It cannot stand alone. Behind it sits a PQR (Procedure Qualification Record), which is the actual test data proving that a sample welded to that recipe meets the required mechanical and inspection criteria. Welders are qualified separately through a WQT (Welder Qualification Test), which certifies the individual operator on the process, positions and materials they are allowed to run. AWS B2.1 describes this chain, and ASME Section IX QW-482 gives you a ready-made WPS format if your customer works to ASME. As a practical reference, this overview of why SMAW remains the shop default matches how most cast iron procedures are actually written, since shielded metal arc welding with nickel-iron electrodes is the process most qualification procedures assume.

Ductile Iron Pipe: The Documented Exception

Pipe work is where the general rules bend. Most weld joint designs used for carbon steel are suitable for ductile iron, but DIPRA’s field welding guidelines state plainly that butt welding of ductile iron pipe is not recommended. The preferred field connections are mechanical joints, flanged joints and special joint assemblies rather than welded butt ends. Preheating ductile iron pipe is normally unnecessary and impractical in the field — you cannot bring a buried or supported line up to 400°F in the rain — though warming the immediate weld area in extreme cold is sensible and helps avoid a thermal shock crack. Field practice is also why portable process selection matters as much on the job site as in the shop.

Safety and Documentation

Nickel-bearing electrodes and cast iron fumes both fall under the ventilation, respiratory protection and training requirements of accepted cast iron welding practice and ANSI Z49.1 as enforced by OSHA. Keep the PQR, the WPS revision, welder continuity records and final NDT reports together for each repair. When a casting comes back years later, that folder — not your memory of the job — is what proves the repair was done to a defensible standard.

When Not to Weld Ductile Iron — Alternatives and Final Best-Practice Checklist

Everything up to this point has covered how to weld ductile iron successfully. This part is about the harder judgment call: deciding not to weld at all. Ductile iron is genuinely weldable and often more forgiving than gray iron, but weldable is not the same as universally repairable. Some castings should be rejected on sight, and some technically possible repairs cost more than the part is worth once preheat, nickel filler, stress relief and machining are counted.

Reject Criteria: Castings That Should Stay Unwelded

Experienced repair shops screen candidates before striking an arc. The usual grounds for refusal:

• Sealer-impregnated castings. Many pump, valve and transmission housings are vacuum-impregnated with plastic or glass sealer to make them pressure-tight. Under arc heat the sealer decomposes, leaving gas paths, porosity and contaminated fusion zones.
• Parts that cracked in service under high restraint. If a casting failed from fatigue or overload while locked into a stiff assembly, those residual and operating stresses remain. Welding adds a new stress field on top, and the crack tends to return a few inches from the weld.
• Safety-critical fatigue components. Brake housings, suspension members, load-bearing brackets and pressure-retaining parts where a hidden defect has no acceptable failure mode.
• Austempered ductile iron (ADI). The engineered ausferritic structure is destroyed in the heat-affected zone. You are no longer welding the material the engineer specified.
• Ni-Hard and other martensitic or white cast irons. Generally unweldable, since high carbon and alloy content make them extremely crack susceptible.
• Unknown grade or history. No markings, no material cert, no service record, or a visible “do not weld” stamp: walk away.

Alternatives That Avoid Fusion Welding

When the base metal must not melt, you have options:

• Braze-welding with nickel-bronze. The filler wets the surface below the casting’s melting point, so there is no fusion zone, no white iron and far less shrinkage stress. This is the classic low-risk repair for thin sections and intricate contours, and the same principle underlies air acetylene torch work used for small localised builds.
• Post-weld annealing. A full anneal can dissolve martensite and restore a ferritic matrix around nodular graphite, increasing deformability of the repaired zone. It works, but adds cost, handling and transport, and research on repair-welded ductile iron confirms the result is not always justified for the part’s service demands (MDPI Metals, 2023).
• Mechanical retention. Sleeves, shrunk-on collars, pins, or machining the bore oversize and fitting a liner. Often faster than a controlled weld and far more predictable.
• Replacement. For commodity castings with short lead times, buying new is frequently the cheapest total cost once setup, fixturing, heat treatment and inspection are counted.
• Returning the casting to the foundry. For large or valuable new castings, the foundry can weld the defect and run the proper anneal under controlled conditions — which is how most foundry weld repairs are handled under AWS D11.2/D11.2M.

Repair Economics: Doing the Math Honestly

Repair welding of defects in new castings is the single largest application of welding iron castings, so procedure discipline directly reduces foundry scrap (Modern Casting). The same arithmetic applies in the repair shop: nickel rod, controlled preheat, slow cooling and post-weld machining are cheap compared with a six-week replacement lead time, but expensive against a $200 off-the-shelf housing. Price the full repair — labour, filler, energy, heat treatment and inspection — against the delivered cost of a new part, then add a risk premium for anything whose failure has consequences. If the numbers are close, replace it.

The 12-Point Ductile Iron Repair Checklist

1. Identify the grade and service stresses; check for “do not weld” markings and reject unknown or sealer-impregnated castings.
2. Clean and degrease thoroughly; degas oil-saturated castings before heating.
3. Remove the defect completely by air arc gouging, grinding or machining — never weld over a suspected crack.
4. Run dye penetrant or magnetic particle inspection on the cleaned cavity.
5. Machine a V or U groove and grind off the casting skin to sound metal.
6. Preheat uniformly at roughly 40 °C (100 °F) per hour; use 200–315 °C (400–600 °F) for most repairs and up to 500 °C for heavy, complex sections.
7. Choose filler deliberately: ENiFe-CI is the default for ductile iron; never exceed 760 °C (1400 °F) base-metal temperature.
8. Butter the joint faces first, then run short stringer beads in alternating positions with strict interpass control.
9. Clean each pass; peen the final pass only, lightly.
10. Slow cool under an insulating blanket, dry sand or in the furnace — never in an air blast or water.
11. Add stress relief or annealing only where the service demands justify the cost.
12. Machine back to specification and finish with final NDT.

Conclusion

The principal answer holds: yes, ductile iron is weldable, and more readily weldable than gray iron — provided you respect its metallurgy. Choose the nickel-iron filler, control the heat, manage restraint, cool slowly, and stick welding remains the forgiving default. But the mark of a real repair specialist is knowing which casting deserves that effort and which should be sleeved, brazed, replaced or sent back to the foundry. Follow the checklist, and the eleven-hour repair will still be intact on the bench the next morning.

Frequently Asked Questions

Can you weld ductile cast iron?

Yes. Ductile iron is generally more weldable than gray cast iron because its nodular graphite dissolves less carbon, producing fewer carbides and less martensite in the heat-affected zone. Success depends on filler choice, controlled heat input, preheat when needed, and slow cooling.

What is the best electrode for welding ductile iron?

A 55% nickel-iron electrode classified AWS ENiFe-CI is the most common default, offering about 50,000 psi tensile strength with low heat input. Use 93% nickel ENi-CI when machinability matters most, and avoid mild-steel electrodes such as E7018 on critical ductile iron repairs because of interface cracking risk.

Does ductile iron need preheat before welding?

Small, unrestrained repairs can often be welded cold with nickel-iron electrodes and short stringer beads. Heavy, complex or highly restrained castings typically need 150–315°C (300–600°F), with furnace preheat up to about 290°C for critical work, heating slowly at roughly 40°C per hour.

Why did my ductile iron weld crack hours after welding?

Delayed cracking is usually caused by hard martensite in the heat-affected zone combined with shrinkage stress and hydrogen in the deposit. It appears as the part cools because the brittle zone cannot accommodate contraction. Fix it with proper preheat, low-hydrogen nickel-iron fillers, short stringer beads, peening and slow cooling.

Can I use E7018 to weld ductile iron?

It can produce strong deposits — about 70,000 psi — but it needs higher amperage and heat input and is far more crack susceptible. A published ductile iron repair procedure lists it only as a secondary option next to ENiFe-CI, and it generally requires post-weld heat treatment to temper the hardened zone.

Is ductile iron weldable to steel?

Yes, transition joints between cast iron and steel are a standard use for nickel-iron consumables. Concentrate the arc centerline onto the mild-steel side so it absorbs more heat, use low current and narrow beads, and preheat the ductile iron side if the section is heavy or restrained.

How do you machine a ductile iron weld repair?

Use a nickel-iron or high-nickel deposit and keep dilution low so white iron does not form at the fusion line. If the deposit or HAZ is too hard, a post-weld stress relief or anneal near 550–600°C will dissolve martensite and restore a machinable ferritic structure.

Should I peen a ductile iron weld?

Yes, light to moderate peening of hot short beads helps stretch the weld metal and offset shrinkage, reducing residual stress. Peen with a round-nose tool using enough force to deform but not rupture the deposit, and clean slag completely between passes.

When should a ductile iron casting be replaced instead of welded?

Skip welding on highly stressed or heavily restrained locations, where the weld would exceed about 20% of the section thickness, on sealer-impregnated castings, or on safety-critical fatigue parts. In those cases brazing, mechanical repair or replacement is safer and often cheaper than a failed weld.