How to Weld Cast Iron to Steel: Step-by-Step Guide, Filler Rods, Preheat & Technique

The bandsaw came in on a Tuesday, wheeled over by a farmer who had been told twice that the frame was beyond saving. A hairline crack ran from a bolt hole in the cast iron body toward the machined table surface, and he wanted it welded back to a steel reinforcing plate he had made himself. “My last guy said cast iron can’t be welded,” he told me. “He said you just heat it and it falls apart.”

That story gets told a lot in repair shops, and it is half true. Cast iron can absolutely be welded, and joining it to steel is routine for anyone who understands the material. What people mean when they say “cast iron can’t be welded” is that it cannot be welded the way steel is welded. Bring the same habits — long beads, high heat, no preheat, a quick air cool — and you will crack it every single time.

This guide is the complete version of how to weld cast iron to steel: what the material does under an arc, which filler metal works, what preheat temperature you actually need (and why the number you read online is probably wrong), the technique step by step, and how to cool the part without undoing everything. It also covers the question every experienced welder eventually asks — whether you should be welding at all, or brazing instead.

Quick answer: Yes, you can weld cast iron to steel. Use a nickel or nickel-iron electrode classified to AWS A5.15 (ENi-CI or ENiFe-CI), preheat the whole casting gradually, aim the arc at the steel side, keep beads short (10–30 mm), peen while hot, and cool the assembly slowly in dry sand or an insulating blanket. Most failed cast iron repairs fail from cooling too fast, not from the wrong rod.

What You’ll Need for the Job

• A stick welder capable of low-amperage DC output — SMAW is the easiest process to control for this work
• ENiFe-CI nickel-iron electrodes for most multipass repair, or ENi-CI (97–99% nickel) for single passes and maximum ductility
• A propane or oxy-acetylene torch for preheat, plus a temperature stick if you want to be precise
• Wire brush, carbide scraper, and grinding wheel for cleaning back to bright metal
• A 3/32 in (2.5 mm) drill bit for crack-stop holes
• A peening hammer — a ball-peen or a blunt chisel edge works
• Dry sand, a ceramic fiber blanket, or vermiculite for slow post-weld cooling
• A magnet, a file, and a bit of patience

Before Anything Else: Identify the Cast Iron

You cannot choose a filler metal or a preheat temperature until you know what you are welding, and “cast iron” is not one material. It is a family, and the members behave completely differently under an arc. Skipping this step is the most common reason a repair fails before the weld even cools.

Grey iron: assume this unless told otherwise

Grey iron is by far the most common casting material you will meet — engine blocks, pump housings, lathe beds, stove bodies, counterweights. Its carbon exists as flakes of graphite, and those flakes are exactly what makes it difficult to weld. If a customer says “cast iron” without qualification, plan for grey iron and its rules.

Ductile (nodular) iron: the forgiving one

Ductile iron, also called nodular or spheroidal graphite iron, has its carbon in the form of round nodules created by magnesium or cerium treatment during casting. Because there are no sharp flakes to act as stress raisers, it is noticeably more weldable than grey iron, especially when the matrix is ferritic. Many ductile iron castings can be welded with little or no preheat at all. A lot of modern pipe, automotive brackets, and agricultural parts are ductile iron even when the owner calls them “cast iron.”

Malleable iron

Malleable iron is white iron that has been annealed so the cementite breaks down into temper carbon. It is tougher than grey iron and generally weldable in a similar way. The subtypes matter: whiteheart malleable has a decarburized low-carbon skin over a high-carbon core, blackheart is more homogeneous at roughly 2.2 to 2.9 percent carbon, and pearlitic malleable is the hardest and least forgiving of the three.

White iron: stop here

White iron holds its carbon as cementite rather than graphite. Engineering references, including TWI, treat it as normally unweldable. It is a wear material — mill liners, slurry pump parts, grinding surfaces — and if your job involves white iron, the honest answer to the customer is a hardfacing repair or replacement, not a fusion weld. Do not accept that work without a specialist conversation first.

A five-minute identification method

Two cheap tests settle most doubts. First, a file test: grey iron files easily and leaves a grey graphite smudge on the file, while white iron is extremely hard and skates. Second, look at a fresh fracture: grey iron breaks dull grey and granular, ductile iron shows some deformation at the edge of the break, and white iron shines silver-white. When all else fails, ask what the part does. A wear surface is a warning sign; a decorative or structural housing usually is not.

One assumption worth killing early: that ductile or malleable iron is harder to weld than grey iron. It is not. The hardest case is the pearlitic one, because more pearlite in the matrix means more martensite forming in the heat-affected zone. Ferritic ductile iron is the easiest cast iron you will ever weld.

Why Cast Iron Cracks Instead of Bending

To understand how to weld cast iron to steel, you first have to understand why the material fights you. Three things are happening at once, and every technique in this article exists to deal with one of them.

Problem one: too much carbon

Cast iron contains more than 2 percent carbon — typically 2.0 to 4.5 percent in grey iron — plus 1 to 3 percent silicon. TWI puts that figure in perspective: it is roughly ten times what you find in most steels. When you put an arc on the material, carbon migrates toward the weld pool and the heat-affected zone, where it forms high-carbon martensite. Carbon is what makes steel strong under controlled heat treatment, and it is exactly what makes a weld zone brittle when it forms martensite by accident. Martensite is hard. Hard things crack.

Problem two: graphite flakes act like pre-scored glass

TWI describes grey iron’s branched, interconnected graphite flakes as forming planes of weakness and acting as stress concentrators that lead to poor toughness. A crack in grey iron does not travel in a clean straight line — it follows the flakes, jumps between them, and creates fresh crack paths as it goes. Think of a sheet of paper that has already been scored: you are not starting the crack, you are finishing it.

Problem three: no ductility to absorb movement

This is the one that decides everything. TWI states it plainly: a ductile metal such as steel relieves stress by stretching, while cast irons have poor ductility, so they crack instead. The stress in a cast iron joint has exactly one outlet, and that outlet is a crack.

One popular myth worth correcting

You will frequently read that grey iron expands at roughly twice the rate of steel, and that this mismatch is why the joint cracks. It is wrong. Published room-temperature coefficients of thermal expansion are close together: grey iron around 11.4 × 10⁻⁶ per °C, ductile iron about 10.6 to 11.2, and A36 steel at roughly 11.7. The twofold figure that circulates in welding articles actually belongs to aluminium, not cast iron. Nickel filler, for its part, sits higher at about 13.3.

Why does this matter practically? Because if you have been cranking up the preheat to compensate for an expansion mismatch that does not exist, you are adding heat you do not need. Cast iron and steel expand at similar rates. The difficulty of this joint comes from low ductility, carbon migration, and the graphite notch effect — not from expansion. Interestingly, nickel’s higher expansion is one reason 55 percent nickel-iron rods produce fewer fusion-line cracks than 99 percent nickel rods: the deposit contracts a little less, so it pulls less on the joint as it cools.

Filler Metal: Why Nickel Works (and What to Pick)

The entire logic of nickel filler rests on one property: it produces a soft, ductile, machinable deposit where steel filler would produce a hard, brittle one that cannot be machined and is nearly guaranteed to crack. TWI explains that a nickel or nickel-alloy consumable produces finely divided graphite, lower porosity, and a readily machinable deposit, avoiding the high-carbon weld metal you get when you weld cast iron with steel filler.

There is a second benefit that matters specifically when you join cast iron to steel. TWI notes that 99 percent nickel gives a soft, malleable deposit, while 55 percent nickel rods are also machinable and — because of their lower coefficient of expansion — produce fewer fusion-line cracks than the pure nickel rod. TWI calls ferro-nickel rods ideal for welding cast iron to steel. That single line is the answer to the most common question in this repair.

The AWS classifications you actually need

The governing specification is AWS A5.15, Specification for Welding Electrodes and Rods for Cast Iron, which covers SMAW, GMAW, GTAW, and oxyfuel consumables. Two exclusions come up in supplier conversations and are worth knowing: AWS A5.14 (nickel and nickel-alloy electrodes) explicitly excludes compositions intended for joining cast irons, and AWS A5.11 defers to A5.15 on the same point. A5.15 also excludes copper-base rods for braze welding cast iron.

Choosing between ENi-CI and ENiFe-CI

Lincoln Electric’s published guidance is the clearest practical version of the decision. Their 99 percent nickel rod is preferred for single-pass welds with high dilution. Their 55 percent nickel rod is preferred for multipass work. Their iron-filler product is used only when the weld must rust like the surrounding cast iron and will not be machined.

That last point is the one people miss, and missing it ruins parts. If the finished casting gets machined, a nickel-type electrode is mandatory. An iron-filler weld on cast iron deposits high-carbon martensite that will destroy a cutter, a boring bar, or a tap the moment it touches the weld.

There is a real tradeoff between the two nickel types. Manufacturer data sheets put 55 percent NiFe deposits at roughly 165 to 218 Brinell and pure nickel around 175 Brinell, and at least one supplier notes that NiFe welds are moderately hard and prefer carbide-tipped tooling. So nickel-iron is softer than an iron deposit but harder than a pure nickel one. If the part will be machined, budget for carbide tooling — or use ENi-CI for the visible or machined passes and ENiFe-CI for the buildup underneath.

Can you use stainless filler instead?

Welding cast iron with E308L or ER308L is widespread shop practice, largely because stainless rod is cheaper and sits in every welder’s locker. Treat it as an unverified shortcut rather than an endorsed procedure. The austenitic deposit is softer and more ductile than an iron deposit, so it shares the underlying logic — but you are working outside the cast iron specification and adding hexavalent chromium to your fume exposure. Prove it on a test piece before you rely on it. For brass or bronze work, see our separate guide on how to weld brass, where filler choices have their own hazards.

Preheat: What the Sources Actually Say

There is no universal preheat temperature for cast iron, and any article that hands you one range as blanket advice is misleading you. The famous “500 to 1,200 °F” figure traces back to Lincoln Electric’s Guidelines for Welding Cast Iron, which states typical preheat temperatures of 500–1,200 °F and warns not to exceed about 1,400 °F because that puts the material into the critical temperature range. That is one school’s method for one set of conditions — an upper-bound approach for large, heavy, or difficult castings, not a recommendation for a small housing repair.

Starting a grey iron repair at 1,200 °F would be a serious error. The casting could be damaged, the whole part would be glowing, and you would be committing to heat you do not need for most jobs.

Preheat by iron type (TWI data)

Manufacturer data sheets run lower than that table in several cases, which is a useful reality check for everyday repairs. One ENiFe-CI product specifies keeping the work below 100 °C. A nickel electrode supplier states that preheating is not required, though 200 °F is acceptable. A cast iron product line suggests 215–230 °F or 250–300 °F with an hour of post-heat. Another ENiFe-CI and ERNi-CI supplier calls for a preheat and interpass temperature of not less than 350 °F.

Where do the very high numbers come from? Iron and steel electrodes on cast iron require preheats around 550 °C to prevent cracking and hard deposits. Those numbers belong to iron-filler and oxyacetylene practice, not to nickel-filler work with modern electrodes. The advice circulates because it was once correct for a different consumable.

The variable that actually matters is cooling rate

Here is the mental shift that improves cast iron results more than any temperature chart: preheat matters because it slows the cooling rate, not because some magic number must be reached. TWI is explicit that heat-affected zone cracking risk is reduced by preheating and slow post-weld cooling, and that preheat should always be applied gradually rather than by blasting one spot with a torch.

Lincoln describes the low-temperature approach cleanly: bring the casting up to about 100 °F, and never heat it so hot that you cannot rest your bare hand on it. That is a genuinely useful sanity check for any cast iron job. If the metal is too hot to touch comfortably, you are in the range where you need to think carefully about what happens next.

How to Weld Cast Iron to Steel: Step by Step

This is the procedure itself. It assumes a stick welder (SMAW) on DC, a nickel-iron electrode, and a stationary repair rather than a production joint. Read the whole sequence once before you strike an arc — the order matters as much as the individual steps.

1. Identify the casting. File test, fracture color, or customer history. Grey iron, ductile iron, and malleable iron each get a different preheat plan; white iron stops the job.
2. Find the full extent of the crack. Cracks hide. Tap the area with a hammer and listen for a dead note, run a dye penetrant or magnetic particle check if the part matters, and mark the crack end-to-end with soapstone. Our guide to soapstone in welding covers layout marks that survive grinding and heat.
3. Drill crack-stop holes. A 3/32 in (2.5 mm) hole at each crack tip gives the crack somewhere to terminate instead of running out into untouched material while you weld. Drill through the crack if it has an end you can find.
4. Vee out the crack. Grind or gouge a V-groove down to sound metal — about 60 degrees included, with a flat root if you can get one. Grey iron’s graphite flakes smear under a grinder, so finish the surface with a carbide scraper or a fresh grinding wheel and then wipe it with acetone. Bright metal, no graphite smear, no oil, no paint, no rust.
5. Dry the casting. A sand casting can hold residual moisture, sand, and binder in the surface. Weld over that and the gas escapes into your puddle as porosity. Warm the whole part slowly before you begin. If porosity keeps showing up on your repairs, read our detailed piece on porosity in welding.
6. Preheat the whole casting, gradually. Heat the entire part, not a local spot — a hot zone next to cold metal is a thermal gradient, and gradients create stress. Roughly 300 °C for ferritic grey iron, 100–150 °C for ferritic ductile iron, nothing at all for many small ductile repairs. Apply the heat slowly and evenly.
7. Set your amperage low. Run on the low end of the electrode’s recommended range. The goal is a small, controlled puddle, not penetration at all costs. Some sources, including TWI and electrode manufacturers, also recommend electrode-negative (straight) polarity to reduce heat-affected zone effects on the cast side.
8. Put the arc on the steel. This is the single biggest practical difference between a repair that holds and one that cracks. Aim the arc at the weld pool and let the puddle wash onto the cast iron — do not dig the arc into the casting. On this joint it is easy: position the steel where the arc is pointing and let the cast iron be the passive partner.
9. Keep the beads short. Lincoln’s STARCAST NiFe instructions call for beads of 10–30 mm; other sources give about 25 mm or 1 inch. Total heat input is heat per unit length times bead length, and short beads with cooling time between them keep the heat-affected zone from sitting at temperature long enough to martensitize. Long stringer beads are efficient on steel and wrong on cast iron.
10. Stitch, skip, and peen. Lay a short bead, stop, and peen it while hot — peening relieves tensile stress as the bead solidifies so the joint is not left pulling against itself. Move elsewhere, come back later: each new segment lands on metal that has already contracted, which spreads distortion instead of concentrating it. When you restrike, reignite on weld metal, never on bare casting.
11. Build up in controlled passes. For a wide or deep groove, butter the cast iron side with a thin layer of nickel-iron first, then fill with subsequent passes. This keeps each individual pass small and keeps dilution under control. For multipass work, ENiFe-CI is the practical filler choice.
12. Control the cooling — this is where the job is won or lost. The moment you stop welding, the work is not done. Wrap the assembly in a ceramic blanket or bury it in dry sand and let it cool over hours, not minutes. No quench, no fan, no cold concrete bench pulling heat from the underside.
13. Finish and inspect. Remove slag, grind the bead flush, and look first along the fusion line — that is where nickel joints fail if they fail. Check for cracks, porosity, undercut, and incomplete fusion. Magnetic particle inspection is a reasonable field check for surface and near-surface cracks; radiographic or ultrasonic testing is more thorough for critical castings.

Why the Cooling Step Deserves Its Own Story

A welder I know once finished a cast iron repair late on a Friday afternoon, wiped the bead clean, and left the part on the steel bench to cool because he wanted to go home. The weld looked perfect. It was still perfect when he left. By Monday morning there was a crack running through the middle of it.

That is a residual stress failure, and it is the most discouraging kind because the weld itself was fine. The metal contracted faster on the outside than the inside, the joint had no ductility to absorb the pull, and the graphite flakes gave the crack somewhere easy to go. Heat input will not fix this. Filler choice will not fix this. Only cooling control will.

The practice that works is simple: the instant the last bead goes down, the casting goes into dry sand or under a ceramic blanket, and the cooling is allowed to take hours instead of minutes. For a large or expensive casting, some welders even leave the furnace set low overnight. The goal is always the same — let the temperature fall gently and uniformly, so nothing in the part is being pulled apart while it still has heat in it.

Six Mistakes That Crack Almost Every Cast Iron Repair

1. Welding like it’s steel. Long beads, high amps, arc dragged across the casting, then a quick air cool. This is the fastest possible route to a cracked part.
2. Using the wrong rod because it was in the rack. E6011, E7018, and plain MIG wire all run on cast iron — and all of them produce a hard, high-carbon deposit that cannot be machined and tends to crack. Nickel filler is the whole reason this repair is possible.
3. Heating one spot with a torch and calling it preheat. A local hot spot next to cold metal is a gradient, and the gradient creates the stress you were trying to avoid. Heat the whole casting, gradually.
4. Not letting it cool slowly. The most common cause of the “it cracked on the weekend” phone call. Plan the cooling before you plan the weld.
5. Skipping identification. White iron cannot be welded, and pearlitic structures behave worse than ferritic ones. Assuming all cast iron is the same is how you end up with a cracked casting you cannot explain.
6. Grinding graphite smear into the joint. Mechanical cleaning smears graphite across the surface, which hurts both fusion during welding and wetting during brazing. Scratch-brush or carbide-scrape, then solvent-wipe.

Brazing Is Often the Better Answer

Before welding a cracked cast iron, ask whether brazing is the right process. On a surprising number of repair jobs it is, and it remains one of the most underused techniques in the modern shop. Braze welding runs at a much lower temperature, so there is far less thermal stress and distortion, the process cracks far less readily, and the finished deposit is soft, ductile, easy to machine, and low in residual stress.

Typical brazing temperatures sit in the 1,150–1,600 °F range, and a braze weld on cast iron needs a preheat of only about 320–400 °C — well below what arc welding demands of the same part. For a decorative housing, a lathe faceplate, a stove casting, or anything where the joint never sees real load or heat, brazing will outlive a fusion weld.

The two limitations you have to check

First, braze-weld strength falls off rapidly above roughly 500 °F in service. That rules brazing out for exhaust manifolds, cylinder heads, turbo housings, and anything else that runs hot. If the joint sees sustained high temperature, you need a fusion weld or a replacement part. Our piece on welding exhaust pipe with a stick welder covers the other side of that temperature problem.

Second, the joint design has to tolerate a filler that is softer than either parent metal — usually an advantage, but not structural steel.

Preparation and filler: where brazing jobs go wrong

TWI warns that mechanical cleaning smears graphite across the surface and that grinding before brazing can prevent the filler from wetting properly. Sandblasting is the better preparation — sandblasted cast iron with a borax flux takes brazing very well, which is why cast iron brazing has been an established trade skill for over a century.

Never use a phosphorus-bearing braze alloy on iron or steel. Johnson Matthey’s guidance is that phosphorus reacts with iron or nickel to form brittle intermetallic phosphide compounds — the joint fails at the interface that was supposed to be its strength. Use a silver brazing alloy covered by AWS A5.31, or a brass or bronze filler matched to the actual service.

MIG, 7018, TIG, and Other “Can I…?” Questions

Can cast iron be welded with a MIG welder?

Yes. AWS A5.15 covers GMAW electrodes, and the matching bare wires are ERNi-CI and ERNiFe-CI. Gas-shielded MIG works well on small, awkward joints where changing sticks is a nuisance, but it is less forgiving than stick about arc control and heat placement — the wire feed keeps putting metal in whether your puddle is behaving or not. For most small repairs, SMAW remains the easier process to control.

Can you weld cast iron with a 7018 electrode?

The arc will run. It is still a bad idea. E7018 is an iron-powder low-hydrogen electrode, and iron filler on cast iron gives you the hard, unmachinable high-carbon deposit and brittle heat-affected zone that cause cracking. If you have already laid a bead with it, expect trouble: grind it out and start over with nickel. There is no technique that makes 7018 a correct answer here.

Do you need to preheat cast iron every time?

No. It depends on iron type, section thickness, filler, and process. Ferritic nodular iron frequently needs nothing. Grey iron generally does. What you are really managing is cooling rate, and preheat is only one tool for it. Keep the whole casting warm, apply heat gradually, and plan the post-weld insulation before you strike the first arc.

Is welding better than brazing for a cracked casting?

Usually brazing, until service temperature rules it out. Below about 500 °F in service, brazing gives lower thermal stress, less distortion, a machinable deposit, and lower cracking risk. Above it, fusion weld with nickel filler — or replace the part. When a customer asks for a weld, it is worth explaining the brazing option; a repair that lasts is better than a repair that was performed exactly as requested.

Cast Iron Types at a Glance

The flake graphite in grey iron is the whole problem: internal stress concentrators, almost no ductility, cracking instead of deforming. Ductile iron moved the carbon into spheres, which is why it is so much more forgiving — and white iron is the exception that matters. If a casting is white iron, the answer is hardfacing or replacement, not fusion welding.

Safety: The Part of Cast Iron Work People Skip

Grinding a cast iron casting is genuinely dangerous

There is a documented OSHA fatality involving a grinding wheel that burst while a worker ground a 250-pound ductile cast iron component. The wheel was running at roughly 3,600 rpm on a wheel rated for about 1,773 rpm. A large cast iron part is heavy, awkward to hold, and it changes the geometry under the wheel as you work — a combination that turns an ordinary grinding operation into a wheel failure. Use the guard, respect the rated wheel speed, and never grind a heavy casting while supporting it by hand.

Fume from cast iron grinding

Grinding and machining cast iron generates iron oxide dust. NIOSH measured iron oxide exposure in cast iron machining and grinding and references a recommended exposure limit of 5 mg/m³ against OSHA’s permissible limit of 10 mg/m³ for iron oxide fume. Chronic exposure is associated with siderosis, a benign pneumoconiosis. NIOSH also gives a respirable graphite limit of 2.5 mg/m³ — relevant, since you are grinding graphite-containing material.

One correction, because it appears in many older articles: cast iron grinding exposure is not a silica hazard. NIOSH attributes the exposure to iron oxide and abrasive material. Silica comes from foundry sand and slag, not from the casting body. If you want the bigger picture on welding fume, our article on whether welding fumes are toxic goes deeper.

Nickel and chromium fume

Nickel and stainless welding fume is treated as an occupational carcinogen concern. NIOSH’s evaluation of stainless welding fume notes that lung cancer risk appears associated with exposure to fumes containing nickel and chromium, and NIOSH has no recommended exposure limit for total welding fume — it recommends controlling exposure to the lowest feasible concentration, which is more stringent than OSHA’s approach. Local exhaust ventilation at the source is the primary control, with respirators only where local exhaust is not feasible.

If you substitute stainless filler for nickel, add hexavalent chromium to your hazard list. OSHA’s permissible exposure limit for chromium VI is 5 µg/m³ as an eight-hour time-weighted average. Note also that OSHA has no enforceable general limit for “welding fume” as a single substance — a 5 mg/m³ limit from a 1989 rule was remanded, so compliance runs through the constituent limits. In practice: ventilate at the source, and don’t weld cast iron in a sealed box. Our welding safety tips guide covers the fundamentals.

Troubleshooting Cast Iron Welds

The delayed crack deserves special attention. A weld that looks fine at the end of the shift and cracks the next morning is almost always residual stress, not a filler problem — and turning up the heat will not fix it. That is the moment when brazing becomes the correct answer.

A Full Repair, Start to Finish

To put the whole method together, here is how a typical cast iron to steel repair runs in practice — a pump housing with a cracked mounting ear that needed to be joined to a new steel bracket.

Assessment (15 minutes): File test confirmed grey iron. Dye penetrant showed the crack extending about 40 mm from a bolt hole. The part would be machined on the mounting face afterward, which ruled out iron filler and stainless, and made ENi-CI the filler for the visible passes.

Preparation (45 minutes): Crack-stop holes drilled at both ends, V-groove ground to sound metal, graphite smear removed with a carbide scraper and an acetone wipe. Steel bracket cleaned to bright metal, fit-up checked, whole casting warmed gently with a propane torch until it was comfortably warm to a bare hand — enough to drive off surface moisture without getting anywhere near the critical range.

Welding (an hour of actual arc time): 1/8 in ENi-CI at low amperage, arc aimed at the steel side, beads of about 20 mm, each one peened hot before moving on. Buttered the cast side first, then filled in three short passes with a rest between each. Total arc time was short; the time between beads was not.

Cooling (overnight): The housing went straight into a bed of dry sand, covered over, and left until the next morning. No fan, no water, no haste.

Finishing: Bead ground flush, magnetic particle check showed no cracks along the fusion line, and the mounting face machined clean with carbide tooling. The customer got a part that worked, and the shop got a repair that did not come back.

Frequently Asked Questions

Can you weld cast iron and steel together?

Yes — it is a well-established repair. Use a nickel or nickel-iron electrode classified to AWS A5.15, keep the arc off the cast iron, use short beads with peening, and slow the cooling as much as you can. For critical joints, work from a qualified written procedure rather than improvising.

What rod do I use to weld cast iron to steel?

For most repair work, ENiFe-CI (55% nickel-iron) is the practical choice — TWI identifies ferro-nickel rods as ideal for the cast iron to steel combination specifically. Use ENi-CI, the higher-nickel electrode, for single-pass welds with high dilution, for the final visible passes, or when maximum ductility matters. If the part will be machined, never use an iron filler.

How hot should I preheat cast iron before welding?

It depends on the iron type. Around 300 °C (570 °F) for ferritic grey iron, 100–150 °C for ferritic ductile iron, and nothing at all for many small ductile repairs. What matters more than the starting number is that heat is applied to the whole casting, applied gradually, and followed by very slow cooling.

Why does my cast iron weld crack the next day?

Delayed cracking is residual stress, not a filler problem. The joint cooled too quickly or was restrained, and the casting had no ductility to absorb the pull. The fix is slower post-weld cooling (dry sand or an insulating blanket), smaller beads, peening, and in some cases switching the repair to brazing.

Is it better to braze than weld cast iron?

Often, yes — below about 500 °F service temperature. Brazing uses less heat, distorts less, cracks less, and leaves a machinable deposit with low residual stress. Sandblast instead of grind before brazing, use a borax flux and a silver or brass filler, and keep phosphorus-bearing alloys away from iron. Above 500 °F in service, fusion weld with nickel filler.

Can I weld cast iron without preheat?

Sometimes. Small, ferritic, ductile iron repairs with nickel filler and good post-weld insulation often succeed cold. Grey iron almost never does on anything beyond a trivial crack. When in doubt, preheat gently and cool slowly — you cannot undo a cracked casting, but you can always have preheated slightly more than necessary.

What is the correct standard for cast iron welding?

ANSI/AWS D11.89, Guide for Welding Iron Castings, is the guide to look at. AWS A5.15 is the filler metal specification covering SMAW, GMAW, GTAW, and oxyfuel consumables for cast iron, and AWS A5.31 covers brazing filler metals. If a job is code-governed, work from the governing code for the application rather than from a guide.

Final Thoughts: Slow Down and Let the Part Tell You What It Needs

Cast iron to steel welding fails for three reasons, and they are almost never the ones people expect: the wrong filler, heat applied in the wrong place, and cooling that was too fast. Nickel filler solves the metallurgy by giving you a deposit that stays soft and machinable. Your real job is thermal — keep the arc on the steel, keep the beads short, heat the whole casting gradually, peen the hot beads, and then wrap the part up and leave it alone for hours.

Two habits will improve your results more than any product recommendation. Learn to read the customer’s part before you reach for the torch, because half the time the joint is quietly asking for brazing instead. And always ask one question before you pick up a rod: does this part get machined afterward? That single answer decides between nickel and iron filler — getting it wrong means the repair can never be finished.

The farmer’s bandsaw frame, incidentally, was welded with ENi-CI in short stitched beads, cooled overnight in dry sand, and machined flat the next afternoon. It is still cutting. The “guy who said cast iron can’t be welded” had simply never been shown the difference between welding steel quickly and welding cast iron carefully.