TIG Welding Cast Iron: A Complete Repair Guide

A cracked exhaust manifold is sitting on the bench, the owner has already been quoted the price of a new one, and someone says, “Just TIG it.” Ten minutes later the arc is humming, the puddle looks fine, and then you hear that faint tick-tick-tick of a crack opening behind the bead. Cast iron is one of the few metals where a beautiful-looking weld can be a failure in progress, and understanding why is the whole game.

TIG Welding Cast Iron: Why This Job Fights Back (and How to Win)

A cracked exhaust manifold is on the bench, the owner has already been quoted the price of a replacement, and someone in the shop says, “Just TIG it.” Ten minutes later the arc is humming, the puddle looks beautiful, and then you hear it: a faint tick-tick-tick somewhere behind the bead as a crack opens in the cooling metal. Cast iron is one of the few materials where a weld that looks perfect can be a failure in progress.

So let’s answer the main question up front. Yes, cast iron can be TIG welded, and it can be welded well. But it only works with the right filler metal, disciplined heat control, and realistic expectations about what the finished repair will look like and how strong it will actually be. Drop any one of those and you are gambling with a casting that is expensive, hard to source, and often load-bearing.

What TIG Gives You on Cast Iron

TIG, or GTAW (gas tungsten arc welding), offers something no other process does: precise, localized heat placement. You can put a small, clean puddle exactly where it needs to go, watch it wet out in real time, and stop before heat spreads into a section you didn’t want to disturb. There is no slag to chip or hide a defect behind, and arc visibility is excellent, which matters enormously when you are chasing a hairline crack into a corner of a gearbox housing or a machine table.

Those qualities make TIG the natural choice for small, high-value repairs where the bead will later be dressed flush and machined. The tradeoff is that TIG offers no forgiveness. The quality of a cast iron weld depends heavily on operator skill because the process has little tolerance for poor heat control. If your arc lingers, if you run too hot, if you let the part cool too quickly, there is no flux coat and no rapid filler deposition to save you. The metallurgy will respond on its own schedule, and it rarely responds kindly.

It is also worth knowing that TWI notes TIG’s highly localized heat input is one reason the process is not generally the preferred method for cast iron, despite producing clean welds. The very precision that makes TIG attractive also creates steep thermal gradients in a material that conducts heat poorly and has almost nowhere to put shrinkage stress. TIG is a scalpel on a job where most welders reach for a blunt instrument.

The Three Variables That Decide Success

Every successful cast iron repair comes down to three decisions, and each one gets its own dedicated section later in this guide:

• Alloy identification. Gray, ductile, malleable, and white iron behave completely differently under the arc. Guessing wrong means the wrong filler and the wrong heat strategy before you have even struck an arc.
• Heat management. You either preheat slowly and uniformly, usually in the 500–1200°F range, or you commit to a deliberate cold-welding technique with short beads and cooling between passes. What you do not do is switch approaches halfway through.
• Filler selection. Nickel-based fillers, roughly 99% nickel or a 55% nickel-iron blend, are the standard. Mild steel and stainless fillers fail because they cannot tolerate the carbon and silicon they pick up from the base metal. We go deeper on rod choice in Alloy Steel Welding Electrodes: Selection, Use & Safety, but for cast iron the answer is firmly in the nickel family.

Skimp on any one of these and the repair may hold for a week, a season, or a single heat cycle before failing in a way that costs far more than doing it right the first time.

What This Guide Covers

We will walk through the metallurgy in plain language, compare TIG against stick, MIG, and oxy-fuel so you know when to reach for a different process, and then go step by step through machine settings, bead sequencing, peening, and slow cooling. After that comes a troubleshooting table for cracks, porosity, hard spots, and ugly beads, followed by real repair scenarios: engine blocks, manifolds, vises, machine tables, and pump housings. Safety, cost, and an honest go/no-go framework round it out, referencing ANSI/AWS D11.89, Guide for Welding Iron Castings, which remains the recognized industry reference for procedures on iron castings.

Along the way we will also be clear about process comparisons. TIG is not always the right tool, and knowing why is just as valuable as knowing how. The same judgment applies in Advantages of Stick Welding: Why SMAW Still Wins, where forgiving electrodes and deep penetration make stick the smarter pick for dirty, out-of-position, or outdoor work.

The goal here is not to make you fearless about cast iron. It is to make you deliberate. Cast iron rewards the welder who prepares slowly, heats with intent, welds briefly, and cools with patience. Hold that rhythm and the tick-tick-tick stays away, and the manifold goes back on the car. Precision TIG work rewards exactly this kind of discipline, and it is a skill worth building the same careful way you would approach Aerospace TIG Welding: Standards, Techniques, and Training. Let’s start with the metal itself.

Know Your Iron Before You Strike an Arc: Gray, Ductile, Malleable, and White

Every welding process has a material that humbles it, and for TIG it is cast iron. The problem is not the torch or the tungsten, it is the metal itself: cast iron is a family of alloys, and the differences between them decide whether a repair holds or cracks before the part cools.

Carbon Is the Whole Story

Most cast irons contain more than 2% carbon, far more than steel. The amount matters less than the shape it takes. In steel, carbon dissolves into the iron lattice. In cast iron most of it separates out as graphite, essentially free carbon with almost no strength. Whether that graphite forms as flakes, nodules, or hard carbides controls ductility, weldability, and how the metal behaves beside a bead.

Gray Cast Iron: The Default, and the Problem Child

Gray iron is the most common cast iron in the world. Engine blocks, vises, machine bases, and old manifolds are usually gray iron. Its graphite forms as flakes that interrupt the metal matrix like thousands of tiny internal notches. Those flakes give it two gifts and one curse: excellent vibration damping, superb machinability, and almost no ductility. It does not stretch meaningfully; it cracks, and thermal contraction stress has nowhere to go before the metal gives up. The flakes also leave the metal microscopically porous, so oil and graphite smears boil into your puddle later.

Ductile (Nodular) Iron: The Weldable One

Ductile iron, also called nodular iron, has spheroidal graphite instead of flakes. A sphere concentrates far less stress than a sharp flake, so ductile iron reaches several percent elongation against the near-zero of gray iron. At the puddle, that difference matters: ductile iron tolerates thermal stress better, cracks less readily, and handles multi-pass work with more patience. Crankshafts, truck components, water pipe, and many modern castings are ductile iron.

Malleable Iron: Small, Tough, and Tricky

Malleable iron starts as white iron, then is heat treated to break the carbides down into irregular clumps of graphite, leaving a reasonably ductile metal used in thin-section hardware, fittings, and small cast components. It is weldable in principle but tricky in practice: the sections are thin and the heat-affected zone embrittles quickly. Thin section plus brittle HAZ equals a narrow window for success.

White Cast Iron: Effectively Off-Limits

White cast iron is extremely hard and brittle because most of its carbon is locked up as carbides rather than free graphite. It earns its keep through abrasion resistance in wear parts, mill liners, and some pump components, and it is essentially unweldable by fusion. Walk away, or braze only if the application truly tolerates it. The reference standard for this work, AWS D11.89, Welding Iron Castings, exists precisely because each of these families behaves differently.

Why the Heat-Affected Zone Bites

The weld itself is only half the problem. The band of base metal just outside the fusion line, the heat-affected zone, is where most cast iron repairs fail. Rapid heating and cooling can form martensite there: a hard, crack-prone structure that appears with no visible warning and may crack hours later. Carbon migration compounds it, as carbon moves from base metal into the weld and the HAZ, raising hardness and brittleness. According to TWI’s technical guidance on welding cast iron, that migration is a primary driver of post-weld cracking, which is exactly why the filler must absorb carbon without turning brittle. Mild steel cannot; nickel-based filler can.

Carbon Equivalent, Thickness, and Restraint

Carbon equivalent estimates how a mix of alloying elements affects hardenability: the higher it is, the greater the chance of hard, brittle structures forming on cooling. Thick sections worsen the problem, holding more heat and pulling harder as they shrink. Restraint matters too, because ribs, thick bosses, and closed shapes lock a casting in place while the weld contracts, concentrating stress at the weakest point. An open flange can give; the ribs cannot, so two identical cracks in two different castings may need completely different strategies.

Practical Identification You Can Do in the Shop

You rarely get a certified material report. Use what the part gives you:

• Spark test: Grind a small area on a stone. Gray iron throws dull red sparks with short, broken trails; white iron throws bright, short sparks; ductile iron produces a slightly longer, bushier stream. A rough test, but fast and free.
• Fracture appearance: Gray iron fractures gray and granular with visible graphite texture; ductile iron usually shows more deformation before it breaks.
• Magnet test: Cast iron is ferromagnetic, so a magnet confirms it is not aluminum or stainless, and the same property powers magnetic particle inspection of near-surface cracks.
• Casting marks: A “65-45-12” or “80-55-06” marking indicates ductile iron grades, while “G” grades usually indicate gray iron.
• Call the manufacturer: For expensive or safety-critical parts, a part-number lookup is the cheapest five minutes you will ever spend.

If the material is still a mystery and the part is anything more than decorative, treat it as gray iron and plan accordingly. Being wrong in the conservative direction costs you some speed. Being wrong in the other direction costs you the casting.

The same habit of knowing your base metal applies across disciplines: our guide on aluminum MIG welding covers similar identification and prep lessons for a completely different set of problems.

Sort out the iron first. Filler choice, preheat strategy, and weld sequence all flow from knowing exactly what you are about to melt, and every other section of this guide assumes you have that answer in hand.

How TIG Behaves on Cast Iron — and How It Compares to Stick, MIG, and Oxy-Fuel

Before you set a single dial, it helps to understand what the arc is actually doing to the metal. Cast iron is high in carbon, low in ductility, and poor at conducting heat away from the point where you strike it. That combination is unusual, and it explains almost every problem you will ever have with this job.

On steel, heat spreads outward from the arc and the surrounding material absorbs the shock of expansion. Cast iron does not cooperate. Heat stays concentrated where the arc lands, the temperature gradient between the arc and the cold surrounding casting gets steep, and when that hot spot shrinks on cooling, the surrounding metal refuses to stretch and accommodate it. The shrinkage stress has nowhere to go, so it goes into the graphite flakes, which act like pre-existing cracks. The result is a bead that looks fine and cracks anyway. The American Welding Society’s standard for welding iron castings, AWS D11.1/D11.1M, is built entirely around managing that reality.

DCEN, AC, and Why Helium Earns Its Place

For cast iron, direct current electrode negative is the default. You get a stable, focused arc, deep puddle control, and predictable heat input. Alternating current is not the automatic answer here, even though many welders reflexively reach for it on difficult metals. AC exists mainly to break up the stubborn oxide skin on aluminum, and cast iron has no such skin to fight. If you want to see what a proper AC setup is actually for, look at a resource like our aluminum prep guide.

The gas is where TIG offers real leverage. Straight argon is fine for thin sections. On thick, heavily restrained castings, an argon-helium mix puts noticeably more heat into the puddle without you stretching arc time and widening the heat-affected zone. Helium-rich shielding also improves wetting, which matters because nickel filler does not flow on cast iron the way mild steel does on mild steel.

Four Processes, Four Different Compromises

Process	Strengths	Limitations	Best suited to
TIG (GTAW)	Precise heat placement, no slag, excellent puddle visibility, bead can be dressed and machined	Slow, unforgiving, demands clean metal and disciplined technique	Small cracks in machined surfaces, thin-wall housings, cosmetic repairs
Stick (SMAW) with nickel electrodes	Forgiving, deep penetration, tolerates dirty or outdoor conditions	Slag removal, coarser bead, less heat control	Field repairs, heavy sections, rough castings
MIG with nickel wire	Fast, economical for production repair, easy arc starts	Heat is less controllable, wire feeds poorly on soft nickel	Repetitive production welding
Oxy-fuel welding and brazing	Broad, gentle heat spreads over a wider area and cools slowly, minimal distortion	Slow, needs fuel gas, fusion welding quality varies	Thin sections, low-stress cosmetic work, historic repairs

Stick deserves more respect than it gets. Nickel electrodes covered by Lincoln Electric’s cast iron guidelines are genuinely forgiving, and stick welding’s advantages in dirty or awkward positions are hard to beat outdoors. MIG with nickel wire speeds up production repair, but the faster deposition comes at the cost of heat control, which is the one thing cast iron punishes.

Oxy-fuel has a long history with cast iron precisely because its heat is broad and gentle rather than pinpointed, and the part cools more slowly behind the torch. Where fusion welding is genuinely risky, technical guidance such as Aufhauser’s cast iron procedures points to brazing fillers like AWS RBCuZn-C and RBCuZn-D instead, keeping the base metal below its melting point entirely. Our air acetylene welding overview covers the torch fundamentals behind that approach.

Where TIG Genuinely Wins

TIG is the right tool when the crack sits in a machined face, when the wall is thin, or when the finished bead has to be milled or ground flush afterward. It is also the process that lets you place heat in a spot the size of a dime, stop, let the part cool, and start again. TIG is a precision tool applied to an unforgiving process, not a shortcut, and matching the process to the repair matters as much as matching the filler to the base metal.

Filler Metal Selection: Nickel 99, Nickel 55, Bronze, and the Traps In Between

If the arc settings are the steering wheel, the filler is the chassis. Get it wrong on cast iron and no amount of technique saves the job. The single most common failure story in this trade is a welder running ER70S-6 into a gray iron manifold because “it’s just steel.” Two weeks later the bead is hard, the HAZ is glass, and the crack is back. Filler selection is where cast iron welding is actually won.

Why Nickel Filler Works at All

The chemistry is straightforward: nickel dissolves carbon without forming the brittle iron carbides that steel filler deposits do. That keeps the weld metal soft and ductile even after picking up carbon and silicon from the base metal. Carbon and silicon are the two elements that travel from the casting into your puddle through dilution, and the filler’s job is to tolerate them. This is exactly why mild steel wire and stainless wire fail here: steel filler gets hard and crack-prone as it carbonizes, and stainless filler forms hard carbides and cracks under restraint. Nickel tolerates the contamination by design.

Filler classification for this work falls under AWS A5.15 for cast iron welding electrodes and rods, with bare nickel TIG wire covered under the nickel alloy filler classifications. TWI’s guidance on cast iron welding makes the same point: filler choice and heat control decide the outcome.

Nickel 99 (ERNi-CI): The Soft, Machinable One

ERNi-CI, commonly sold as Nickel 99, is nearly pure nickel. The stick electrode equivalent, ENi-CI, is a nearly pure nickel electrode known for soft, easily machined deposits and low cracking risk in gray iron repairs. In practice that means:

• Very high ductility, so it absorbs shrinkage strain instead of resisting it.
• Excellent machinability; you can tap, bore, and face the deposit with ordinary tooling.
• High tolerance for carbon pickup, which makes it forgiving on old, uncertain castings.
• Lower strength and a pale, yellowish color match that stands out on a painted or aged surface.

Use it for gray iron, cosmetic repairs, and anything that gets machined afterward.

Nickel 55 (ERNiFe-CI): The Strong, Crack-Resistant One

ERNiFe-CI, sold as Nickel 55, contains roughly 45% iron alongside the nickel. The stick version, ENiFe-CI, is a nickel-iron electrode with that same iron content, giving higher strength, better dilution tolerance, and improved resistance to thermal cracking. The extra iron also gives a darker, more convincing color match on ferrous castings.

Reach for Nickel 55 when you are welding ductile iron, building up multi-pass beads, joining thick sections, or welding cast iron to steel. It handles higher restraint and repeated thermal cycling better than 99% nickel.

Bronze and Copper-Based Options

Silicon bronze and aluminum bronze are not fusion fillers for structural cast iron work, but they earn a place on the rack. They run at lower heat input, wet nicely on prepared gray iron, and are excellent for low-temperature cosmetic builds and non-pressure patches. The limits are clear: lower strength, no metallurgical bond that carries real load, and they are inappropriate anywhere pressure integrity or structural strength matters. Bronzing is a repair of convenience, not a repair of record.

Diameter, Preparation, and Stick Notes

Match filler diameter to section thickness. Thin-wall housings and small cracks take 1/16-inch rod, because it lets you add metal in small, controlled increments. Heavy, restrained castings take 3/32- or 1/8-inch rod to build a bead without excessive passes. Keep filler clean: store rods dry, wipe them, and never let them sit in a contaminated rod holder. For stick work, ENi-CI and ENiFe-CI fluxes are formulated for the process and generally need no additional flux; adding extra flux to a TIG puddle is never correct.

For a broader look at rod and electrode logic, our All Position Welding Rod Guide: Types, Uses & Tips covers the selection habits that transfer to this work.

Filler Decision Table

Casting / Goal	Filler	Heat Strategy
Gray iron, machined surface	Nickel 99 (ERNi-CI)	Cold or light preheat
Ductile iron, multi-pass	Nickel 55 (ERNiFe-CI)	Preheat preferred
Cast iron to steel	Nickel 55	Preheat, slow cool
Cosmetic, non-pressure build-up	Silicon bronze	Low heat input
Thick, highly restrained part	Nickel 55, larger diameter	Hot method, 500–1200°F

Lincoln Electric’s cast iron guidelines reinforce the pairing: nickel fillers for fusion welding, with heat strategy chosen to match the casting, not the other way around.

Preparation: Finding the Crack, Cutting It Out, and Deciding on Preheat

Here is an uncomfortable truth about cast iron: the welding itself is maybe 20 percent of the job. The rest is finding where the metal actually failed, removing everything compromised, and deciding how you will manage heat before you strike an arc. Get preparation right and even an amateur nickel bead will hold. Skip it and a professional with a top-tier TIG rig will produce a crack that shows up three days later on the customer’s bench.

Finding and Mapping the Crack

Cast iron cracks are liars. The visible line is often just the part that reached the surface, so your first job is mapping, not welding. For clean, machined castings, dye penetrant testing is the standard: clean the surface thoroughly, flood it with penetrant, let it dwell as directed (usually 5 to 30 minutes), wipe it off, then dust on developer. The dye bleeds back out along the crack and gives you a pink roadmap, including hairline branches you would never see with the naked eye. TWI’s guidance on welding cast iron puts crack detection and removal first for exactly this reason.

For rough sand-cast surfaces where penetrant runs get swallowed by texture, magnetic particle inspection works well on ductile and malleable irons, though graphite flakes in gray iron scatter the field and muddy the indications. And then there is the oldest trick in the shop, which still works: heat the casting gently or soak it in kerosene, wipe it dry, and dust the surface with French chalk or ordinary whiting. As oil creeps out of the crack over the next hour, it draws a dark line exactly where the metal is broken.

Cutting It Out: Stop-Drilling and Grooving

Once you know the extent, commit to removing all of it. Start by stop-drilling each crack tip, drilling about 1/8 inch (3 mm) past where penetrant shows the crack ending, so the tip is blunted and cannot run further under stress. Then grind or gouge the crack out entirely down to bright, sound metal. Air carbon arc gouging is faster on heavy sections, but TIG torches double as handy grinders on small jobs and give you more control near machined edges.

Open the cavity into a V-groove with a narrow included angle, typically around 60 to 75 degrees. This sounds counterintuitive, but a wide-open groove means a wider bead, more filler, more heat, and a bigger heat-affected zone on a material that punishes exactly that. On thick sections, a U-groove gives good access while still keeping deposited volume low. When in doubt, grind a little deeper and add another pass rather than flooding one giant bead.

Cleaning: The Part Everyone Skips

Freshly ground cast iron still hides a carbon and graphite-rich smear, and castings that have spent decades sucking up oil and coolant are worse. Welding over oil-soaked iron is futile: the porosity you get is not a technique problem, it is a contamination problem, and it will reappear no matter how carefully you rerun the bead. Degrease with acetone, grind the groove walls again after degreasing, and accept that porous castings may need two or three grind-and-clean cycles before the metal stops weeping black.

Managing Restraint

A cracked casting is a spring trying to open, and everything you do either fights that or feeds it. Fixture the part so the joint is closed and aligned before welding, and think through your bead sequence before lighting up: backstep patterns, alternating sides of the groove, and short staggered passes distribute shrinkage stress instead of piling it into one spot. For long cracks in rigid, heavily restrained castings, mechanical solutions earn their keep: plug welds over cross-drilled holes, steel keys fitted into machined slots across the crack, or stitched welds with unrepaired gaps between segments. Some large or complex cracks are honestly better brazed or pinned than fusion welded.

Preheat: Hot Method or Cold Method

Now the big decision, and it must be made before the first pass. Lincoln Electric’s cast iron guidelines give typical preheat temperatures of 500 to 1200 degrees Fahrenheit and warn against exceeding roughly 1400 degrees, which pushes the iron into the critical range where fresh hard phases form on cooling. Preheating must also be slow and uniform. Torch-spot-heating one area of a large, complex casting can crack the part all by itself as sections expand at different rates. Use an oven, a furnace, or a bed of insulating blankets, and soak the whole part evenly.

The alternative is the deliberate cold method: no preheat at all, beads of roughly one inch, peening each bead, and letting it cool before the next. That approach trades speed for distortion control, which is why it suits machine surfaces. What you must not do is start cold, get anxious halfway through, and decide to preheat the remaining half. Mixing strategies mid-job creates a steep thermal boundary and a very predictable crack. Whichever route you pick, commit to it.

Measurement matters on both paths. Temperature sticks (paint-like crayons that melt at set thresholds) are cheap insurance, and infrared guns give instant readouts, but read the manual’s emissivity note first. Oxidized, dark cast iron reads close to true; a freshly ground shiny patch or a greasy surface will fool the reading, sometimes by a couple hundred degrees. Aim the gun at a representative oxidized area, or use a temperature stick to verify the gun once per job. Aufhauser’s cast iron procedures also note that when no post-weld heat treatment follows, the casting should cool slowly under an insulating material such as lime or vermiculite, and that stress relieving near 1150 degrees followed by furnace cooling to at least 700 degrees is the recognized treatment for castings that must return to service under load. All of that starts with the prep decisions made right here, before the arc ever strikes.

The TIG Procedure Step by Step: Settings, Technique, Peening, and Cooling

Enough theory. The hood goes down. This sequence assumes you have identified the iron, mapped the crack, grooved it out, and committed to either the hot or the cold method. What follows is the hands-on order of decisions, from machine settings to the moment the part disappears into the blanket.

Machine Setup Before the Arc Starts

Set the machine to DCEN (direct current electrode negative), the default for TIG on steel and cast iron alike: most of the heat lands in the work, the arc is stable, and tungsten life is good. AC is not the automatic answer here. AC exists largely to break up aluminum’s stubborn oxide skin, a job covered in guides like Aluminum Cleaner for Welding: How to Prep Aluminum Right, and cast iron has no such layer to scrub.

Use a 2% lanthanated or thoriated tungsten, ground to a sharp point. Do not ball the tip; a balled end is for AC on aluminum and it blunts your arc on thin cast sections. Flow 15–20 CFH of argon through a standard cup, or 12–15 CFH with a gas lens, which buys wider, calmer coverage in tight grooves. Gas flow cuts both ways: too little invites atmospheric contamination, while excessive flow creates turbulence that pulls air into the arc, a common TIG defect cause noted by Miller. For thick, heavily restrained sections, an argon-helium mix adds heat to the puddle without you forcing amperage. Broader torch and machine fundamentals are covered in Aerospace TIG Welding: Standards, Techniques, and Training.

Amperage Starting Points (A Tuning Table, Not Gospel)

Start here, then adjust by watching how the puddle behaves:

Base thickness	Filler diameter	Starting amperage
1/8 in (3 mm)	1/16 in (1.6 mm)	60–80 A
3/16 in (5 mm)	3/32 in (2.4 mm)	80–110 A
1/4 in (6 mm)	1/8 in (3.2 mm)	110–140 A
1/2 in (12 mm)	1/8 in (3.2 mm) + helium	140–180 A

Err on the cool side. If the puddle wets out and flows without glowing aggressively, you are in range. If it churns or the edges boil, back off and slow down rather than adding current.

Technique: Wet Out, Don’t Burn In

Establish the puddle on the base metal, not on the filler rod. Once it wets, dab the nickel rod at the leading edge and advance a little at a time. Keep the arc length tight, roughly one tungsten diameter, so heat stays where you put it. On gray iron, a shiny, fluid puddle that creeps outward is correct; one that spits or darkens means contamination or too much heat. Stop the bead while the puddle is still healthy, well before heat spreads into the surrounding casting.

Bead Sequencing, Peening, and Crater Filling

Work in short passes of about 1 in (25 mm), using a backstep pattern or alternating sides so shrinkage stress cancels itself instead of stacking up. Let each pass cool to the touch before the next one. TWI describes preheat and interpass control as the single most important factor in avoiding stress cracking in cast iron, and interpass temperature is where that discipline lives.

After each bead, peen lightly along the weld metal with a round-nose chisel while it is still dull red, relieving contraction stress. Peen the bead, never the surrounding casting, and never hard enough to smear or cold-work the parent metal. Finish every pass with a proper crater fill: taper the current or step back into the puddle and add filler. Ending abruptly lets the puddle shrink and freeze unfilled, and crater cracking is a well-documented TIG failure mode described by Miller.

Cooling and Post-Weld Care

Insulate the part the moment the final crater is filled. For critical work, controlled stress relief follows the guidance in Lincoln Electric’s cast iron welding guidelines: a slow reheat to around 1100°F followed by furnace cooling. Once cool, clean the bead with a wire brush. Never quench it, and never hit it with a sudden blast of shop air.

Worked Example: A Cracked Mounting Flange

1. Dye-penetrant the flange face, mark the crack end to end, and stop-drill each terminus 1/8 in.
2. Grind a narrow-V groove, about 45° included angle, through the full crack to bright metal; degrease twice.
3. Preheat the whole flange uniformly to 500°F with a torch, confirming with a temperature stick.
4. Weld three 1 in backstep beads with ERNiFe-CI 3/32 rod at roughly 90 A, peening each bead and letting it cool to touch.
5. Fill the final crater, then bury the flange in dry sand for slow cooling.
6. Dress the face with a carbide cutter, check for a file-tested soft bead, and bolt it up without forcing.

Do that sequence consistently and the arc stops being a gamble. Where distortion truly cannot be tolerated, precision alternatives like Alpha Laser Welding: Systems, Uses, and Safety Guide are worth understanding. For heat control across processes, our Advantages of Stick Welding: Why SMAW Still Wins guide explains why a stick nickel electrode remains the forgiving fallback when conditions are less than ideal.

Troubleshooting: Cracks, Porosity, Hard Spots, and Ugly Beads

Cast iron has a way of surprising you. The bead looks clean, the part cools properly, and a hairline crack shows up two days later on the bench. Saving the part instead of scrapping it comes down to diagnosis: which defect you actually have, what caused it, and whether the casting is still salvageable. This is the reference you want open while the iron is still cooling.

Defect	Most Likely Causes	Fix
Immediate cracking in or beside the bead	Excessive heat input, wrong filler, high restraint, no preheat on thick sections	Grind to sound metal, re-preheat, shorter beads, peen while hot
Delayed cracking (hours or days later)	Rapid cooling, martensite in the HAZ, contraction stress, no peening	Re-grind, stitch weld, peen each bead, insulate for slow cooling
Porosity or pinholes	Oil, moisture, graphite contamination, poor gas coverage, turbulent flow	Grind to bright metal, dry the material, improve shielding, re-weld
Hard spots / unmachinable zones	Carbide and martensite from overheating, wrong filler, too much heat per pass	Preheat discipline, less heat per pass, carbide tooling, light cuts
Arc wander, poor wetting	Magnetized casting, dirty surface, arc length too long	Demagnetize, clean thoroughly, shorten arc, consider AC
Tungsten inclusions	Tungsten touched the puddle or filler	Grind out completely, replace, widen technique
Color mismatch, dull or yellowish bead	Inherent to nickel fillers, especially 99% nickel	Switch to nickel-iron filler or plan on machining and painting

Immediate and Delayed Cracking

Cracking is the defect that earns cast iron its reputation. TWI’s guidance on welding cast iron points to the same culprits every time: a hard, brittle heat-affected zone combined with contraction stress the base metal has no ductility to absorb. Immediate cracks usually mean too much heat went in at once, or the filler could not tolerate the carbon picked up from the base metal. Delayed cracks are more sinister because the repair looked successful; they come from rapid cooling that leaves martensite and locked-in stress behind.

The fix is the discipline covered earlier in this article, applied without exception: grind the crack out completely to bright metal, keep beads short, peen each bead with a round-nose chisel while it is still hot, and stitch long cracks rather than running one continuous pass. On highly restrained castings, alternating sides of the groove and letting the part cool to the touch between passes does more for crack prevention than any amount of amperage tweaking. Lincoln Electric’s cast iron guidelines emphasize controlling heat input and cooling rate above almost everything else, and that advice holds up in the shop.

Porosity: The Contamination Trap

Per ESAB’s technical guidance, porosity in TIG welds is caused by entrapped hydrogen, nitrogen, oxygen, or moisture, and it is made worse by contaminated base material. Cast iron is a worst-case substrate. Its graphitic, porous structure acts like a sponge for oil, coolant, and grease, and those contaminants outgas under the arc, often producing bubbles that only appear after the puddle has frozen. A groove that looked clean can still bleed oil from the pores behind it.

The practical answer is aggressive mechanical cleaning: grind the groove, degrease, heat the surrounding area to drive out absorbed oil, then grind again. Poor gas coverage and turbulent flow add to the problem, so check your cup condition, keep the arc length tight, and shield the still-hot bead during post-flow. Expect to grind out and re-weld porosity at least once on oil-soaked castings; that is normal, not a sign you are failing.

Hard Zones and Machinability

A hard heat-affected zone shows up when you take a file to the weld and it skates instead of cutting. Carbide formation and martensite are the causes, and AWS D11.89 addresses these risks through controlled heat input and nickel-based filler procedures. The simple shop test is the file: if it will not bite, plan for carbide end mills and light cuts, because a hard HAZ can blunt high-speed steel tooling almost instantly. Prevention is cheaper than remediation here. Less heat per pass, proper preheat, and a filler matched to the casting type keep the HAZ soft enough to machine.

Arc Wander, Wetting Trouble, and Tungsten

An arc that dances sideways across the groove is often a magnetized casting, common with iron that has been near motors or magnetic chucks. Demagnetize before you strike an arc. If that is not possible, a slight AC bias can tame the wander, though Miller’s common TIG problem guide notes that poor wetting and erratic arcs more often trace back to dirty base metal and arc length that is simply too long. Tungsten inclusions have one cause: the electrode touched the puddle or the rod. Grind the contaminated tungsten out entirely; you cannot leave a tungsten inclusion in a cast iron repair and hope the next pass covers it.

Ugly Beads and Realistic Cosmetics

Be honest with yourself and the customer about appearance. Nickel fillers run differently than steel, wet out wider, and leave a bead that rarely looks like clean steel TIG work. Discoloration and rough surface texture are expected, not defects. Judging cosmetic quality matters on visible repairs too, much like the decisions covered in alloy wheel repair planning, where the finish requirement drives the whole process choice.

Prevention Checklist

• Identify the cast iron type before selecting filler or heat strategy.
• Grind every groove to bright, oil-free metal and re-clean if staining bleeds out.
• Match filler to casting type and repair goal; never use mystery rod.
• Commit to either hot or cold method and stay with it through cool-down.
• Keep beads short, peen while hot, and stitch long cracks.
• Insulate for slow cooling; never quench or set the part on a cold steel bench.
• File-test the weld zone before machining; switch to carbide tooling if it is hard.
• Inspect with dye penetrant after cooling, not before.

Most failures trace back to one skipped line on that list. If a repair keeps cracking after two honest attempts, stop and reconsider the process or the economics. AWS D11.89 exists because cast iron welding rewards procedure and punishes improvisation, and a repair done to documented practice is one you can defend.

Real Repairs: Engine Blocks, Manifolds, Vises, and Machine Parts

Technique only matters once you can read the job in front of you. Two castings can look identical on the bench and call for opposite strategies: one wants preheat and slow cooling, the other wants to stay cold and short-beaded. Here are five repairs you will actually meet, with the verdict, filler, heat plan, and failure modes for each.

Cast Iron Exhaust Manifold: The Worst Common Job

Exhaust manifolds are among the most demanding cast iron repairs you can take on. The casting has spent years heat-cycled from ambient to exhaust temperature, the sections are thin, and the shape is rigid, so repeated thermal cycling concentrates stress right where you intend to put a bead. As TWI notes, cast iron is inherently difficult to weld because of its high carbon content and low ductility, and a heat-cycled manifold amplifies both problems.

Verdict: replace or braze first, TIG only when the crack is in a non-critical area such as a mounting flange. Filler: ERNiFe-CI (Nickel 55), which handles the repeated cycling better than 99% nickel. Heat: preheat the whole manifold uniformly, then cool it buried in dry ash or a blanket. Watch for: cracking in the next heat cycle, and warping that leaves the flange leaking even when the bead is sound.

Engine Block: Coolant Passage or Mounting Ear

Cleaning is the brutal part. Oil and coolant soak into the porous surface for hundreds of thousands of miles, and grinding that stops short of bright metal just pushes contamination back into the puddle. Groove the crack fully, degrease, then regrind, and repeat until the metal comes up clean and shiny. Lincoln Electric’s cast iron guidelines stress both thorough cleaning and controlled heat input as the twin requirements.

Verdict: repairable if you can reach the crack with a grinder. Filler: Nickel 55 for anything structural, Nickel 99 for a machined sealing face. Heat: preheat where the block allows it; a fully dressed block often will not, which pushes you toward the cold method with strict pass discipline. Watch for: porosity that keeps reappearing, and a hard spot that chews a carbide end mill on the first machining pass.

Antique Bench Vise: Thick, Gray, and Stubborn

Most old vises are gray iron in heavy sections with high restraint, which is exactly the geometry that traps shrinkage stress. High-restraint castings with intersecting sections crack far more readily than simple open shapes because the shrinkage has nowhere to relieve itself. This is a textbook candidate for preheat plus Nickel 99 and very slow cooling. Manage the cosmetics: a 99% nickel bead will not color-match gray iron, and it will not until it has been dressed.

Machine Table or Gearbox Housing: Distortion Is the Enemy

When machined surfaces must stay flat, heat is your adversary rather than your tool. Preheating a large machine frame is often physically impossible, and preheating the entire casting is preferred where practical, which in practice favors cold welding for big frames. Use short beads, alternate sides, peen each bead with a round-nose chisel, and let the part cool to the touch between passes. A AWS D11.89 procedure is the right reference for anyone doing this repeatedly in a production setting.

Pump and Valve Housings: Pressure Integrity Rules

For pressure-retaining castings, a bead that merely looks good means nothing. Repair with Nickel 55, then hydrostatic or leak test before the part goes back into service. If you cannot test it, you should not sign it off.

Repair	Verdict	Filler	Heat strategy	Failure mode
Exhaust manifold	Braze/replace first	Nickel 55	Full preheat, slow cool	Recrack on heat cycle
Engine block passage	Repairable if reachable	Nickel 55/99	Preheat if possible	Porosity, hard spots
Bench vise	Good candidate	Nickel 99	Preheat, buried cool	Restraint cracking
Machine table	Cold weld preferred	Nickel 55	No preheat, short beads	Distortion
Pump/valve housing	Repair, then test	Nickel 55	Job dependent	Leakage

The Economics, and the Honest No

Filler rod is the cheap part. The real cost is labor: crack mapping, groove grinding, preheat, slow passes with cooling between them, peening, insulation, machining, and inspection. For a manifold that costs less than a day’s labor to replace, welding is a losing proposition no matter how well you weld it. A technically successful bead on a part that should have been scrapped is still a failed repair, and saying so is part of the job. For a rare casting that has not been made in forty years, the arithmetic flips, and the same hours become the only sensible option.

Safety, Cost, and Knowing When to Walk Away

Every part of a cast iron repair happens at the edge of the material’s tolerance, and that goes double for the person holding the torch. The fumes are worse than on mild steel, the iron stays hot far longer than it looks, and the economics often argue against the repair before you ever light up. Let’s deal with all three honestly.

What’s Actually in the Air

Nickel-based fillers are the backbone of cast iron TIG work, and nickel is also a recognized occupational respiratory hazard. Nickel and nickel compounds are linked to respiratory sensitization and, with long-term overexposure, to cancer of the nasal cavity and lung. That means welding ERNi-CI or ERNiFe-CI without local exhaust ventilation is not a small lapse. A movable fume extractor at the arc, or a supplied-air respirator in tight spaces, is the baseline.

• Ozone and nitrogen oxides: TIG’s intense ultraviolet light converts oxygen and nitrogen in the air into ozone and NOx. These irritate eyes, throat, and lungs, and they build up fastest in a still, confined shop. OSHA’s welding and cutting overview and the NIOSH welding topic page both treat this as a core exposure concern.
• Metal fume from dirty castings: old engine blocks carry oil, coolant residue, paint, and sometimes leaded or zinc-bearing coatings. Grinding and welding all of it aerosolizes whatever is there.
• Shielding gas displacement: argon and helium are heavier than air in a pit or tank. In an enclosed space they can push oxygen below safe levels without any warning smell. Never weld castings inside a vessel without atmospheric monitoring and an attendant.

OSHA 1910.252, the general industry standard for welding, cutting, and brazing, sets the ventilation, fire prevention, and protective equipment requirements that apply here. Read the ventilation table before you assume a shop fan is enough. It usually isn’t.

Fire, Torch, and Oven Discipline

Preheat means open flame or glowing elements, often for hours. Clear the area of oil rags, cardboard, and solvent cans before the oven comes on. Keep a dry chemical extinguisher within reach, know where the gas shutoff is, and never leave a torch lit and unattended. If you’re heating with a rosebud, check for combustible dust on the floor around the casting, because cast iron grinding dust plus a stray spark is a genuine hazard.

The Real Cost of a “Cheap” Repair

Filler is often the smallest line item. Nickel rod runs several times the price of mild steel per pound, and a deep groove in a thick flange can eat a surprising amount. The bigger costs are time and uncertainty:

Cost Item	What to Expect
Filler and consumables	Modest per repair, but multiplies on multi-pass work
Preheat and controlled cool-down	Hours that cannot be compressed without risking cracks
Machining after weld	Often required for sealing surfaces; nickel 99 helps here
Inspection and testing	Dye penetrant, pressure test, or both
Rework risk	The wildcard that sinks the economics

Compare the total against a replacement part before you commit. For a $90 aftermarket water pump housing, the math rarely favors welding. For a discontinued antique vise or a matching-numbers engine block, it absolutely does.

A Go/No-Go Framework

1. Structural or cosmetic? Cosmetic repairs forgive a lot. Load-bearing repairs on restrained castings need a real procedure and inspection.
2. Restrained or free? Intersecting ribs and thick flanges trap shrinkage stress. Free edges move and survive.
3. Thick or thin? Thin sections preheat poorly and distort; very thick sections need serious heat input and long cool-downs.
4. Clean or oil-soaked? If you cannot grind to bright, dry metal, stop. Porosity will find every pocket.
5. Code-governed? If the part falls under a standard, follow it and document it.

Pressure vessels, lifting lugs, trailer hitches, and anything whose failure could injure someone belong with a qualified repair shop that has written procedures, AWS D11.8 guidance on hand, and real testing capability. A tidy bead is not a qualification. If you want to deepen your general process knowledge before tackling iron, our aerospace TIG welding guide covers arc control and procedure discipline that translate directly to precision cast iron work.

The honest answer, sometimes, is no. Saying it out loud protects your customer, your insurance, and your reputation far better than a weld that fails six weeks later.

Conclusion: A Practical Checklist for TIG Welding Cast Iron

By now you can see that TIG welding cast iron is not really an arc-handling problem. The puddle work is ordinary GTAW. What decides the outcome is everything wrapped around it: knowing what metal you have, where the heat goes, which rod you dip, and how the part is allowed to cool. TWI puts it plainly — successful cast iron welding depends on identifying the alloy, controlling heating and cooling, and choosing a suitable process. Get those three right and the arc takes care of itself.

The Decision Chain, Start to Finish

Print this out and keep it near the machine. Every step feeds the next one, and skipping any of them is how cracks start.

1. Identify the iron. Gray, ductile, malleable, or white. Use casting marks, fracture appearance, spark test, or a magnet. Wrong ID means wrong filler and wrong preheat.
2. Inspect and map every crack. Dye penetrant or magnetic particle for critical parts, the oil-and-whitening trick for rough castings. Find the ends before you strike an arc.
3. Decide repairable or replaceable. Answer this honestly before any metal is cut. Pressure parts, lifting lugs, and safety-critical castings belong in a qualified shop.
4. Choose the filler. Nickel 99 for gray iron, cosmetic work, and machined surfaces; Nickel 55 for ductile iron, multi-pass builds, and cast-iron-to-steel joints.
5. Pick one heat strategy and stay with it. Lincoln Electric’s guidance is to choose either the hot method or the cool method and use it for the entire repair. Switching halfway through is worse than either approach alone.
6. Prepare the groove. Stop-drill the crack tips, grind to bright sound metal, open a narrow V, degrease, and remove every trace of graphite and oil.
7. Weld in short controlled passes. Backstep, alternate sides, and keep beads short with cooling intervals between them — short welds with cooling pauses cut heat buildup and residual stress compared with long continuous beads.
8. Peen the bead only. Round-nose chisel on the deposited metal, never on the base casting, while the bead is still hot.
9. Cool slowly. Blanket or dry sand, no drafts, no quenching, no compressed air. Let it come down to touch temperature on its own.
10. Inspect and finish. Re-check with penetrant, run a file over the HAZ to catch hard spots, then machine, lap, or pressure-test as the part demands.

The Three Non-Negotiables

If you remember nothing else from this guide, remember these:

• Clean base metal. Oil, coolant, paint, and graphite are baked into cast iron’s pores. Contamination shows up later as porosity, and porosity shows up as a second repair.
• A nickel-based filler. Mild steel and stainless rods cannot tolerate the carbon and silicon they pull out of the base metal. The weld simply cracks.
• Disciplined heat. Preheat uniformly or weld deliberately cold, peen, and cool slowly. Thermal stress is the enemy, and it never stops working on you.

One last piece of perspective. TIG is a precision tool applied to an unforgiving process — not a shortcut. A weld that looks beautiful and holds no load is still a failure, and a technically perfect repair on a casting that should have been scrapped is a waste of your time and the customer’s money. Knowing when to walk away is part of the skill.

So practice first. Buy a broken gray iron housing or a scrap engine part, crack it deliberately, and run the full sequence end to end: identify, map, groove, preheat, bead, peen, cool, inspect. Do that three or four times and the checklist stops being text on a page and starts being habit. When you’re ready to go deeper, our guides on filler and electrode selection and TIG technique and setup standards cover the details that will sharpen the rest of your work.

Frequently Asked Questions

Can you TIG weld cast iron successfully?

Yes, but it is a controlled repair rather than a routine weld. TIG works best on cast iron when you use a nickel-based filler, clean the crack out completely, and manage heat with either a proper preheat or short, cool, peened beads. The result is usually a serviceable repair rather than a restoration to original strength.

What filler rod should I use for TIG welding cast iron?

Nickel-based rods are the standard choice. Nickel 99 (ERNi-CI) gives a soft, machinable, ductile deposit that suits gray iron and cosmetic repairs, while Nickel 55 (ERNiFe-CI) is stronger and better for ductile iron, thicker sections, multi-pass work, and joining cast iron to steel. Avoid mild steel and stainless fillers because they form brittle carbides.

Do I have to preheat cast iron before TIG welding?

Not always, but you must choose one strategy and stick with it. Preheat of roughly 500–1200°F slows cooling and reduces stress, which suits thick, restrained, or highly stressed parts. The alternative is cold welding with short beads, light peening, and full cooling between passes. Switching between the two mid-repair invites cracking.

Why did my cast iron weld crack after it cooled?

Cracking is usually a stress problem, not a technique problem. Carbon pickup hardened the weld or heat-affected zone, the part was restrained, the heat input was too high or too concentrated, or cooling was too fast. Delayed cracking that appears hours later is common and is why slow cooling and, where possible, stress relief matter.

Can TIG welding cast iron be machined afterward?

Nickel 99 deposits machine reasonably well, which is one reason it is popular for repairs on machined surfaces. Nickel 55 is harder and tougher to cut. The bigger issue is often the heat-affected zone in the casting itself, which can harden enough to require carbide tooling and light cuts.

Is TIG better than stick welding for cast iron?

They solve different problems. TIG gives precise, slag-free, controllable heat for small and visible repairs, but it is slow and concentrates heat. Stick with a nickel electrode is more forgiving on dirty or awkward castings and puts down metal faster. Many shops keep both options available depending on the job.

Can you weld cast iron to steel with TIG?

Yes, and a nickel-iron filler such as Nickel 55 is usually the best choice because it tolerates dilution from both sides of the joint. Keep the heat balanced, avoid weaving, and expect the joint to be strong but not identical to a steel-to-steel weld in fatigue performance.

What castings should not be welded?

White cast iron is generally considered unweldable by fusion because of its carbide structure, and thin, complex, highly stressed castings often crack regardless of technique. Safety-critical parts such as lifting components, pressure vessels, and brake or steering parts should be replaced or repaired by a qualified shop with proper testing.

How do I clean a cast iron crack before welding?

Grind the crack out completely to sound metal, opening a narrow groove with a burr or thin wheel, and stop-drill the ends first. Then degrease aggressively, because cast iron is porous and holds oil. Repeat grinding after the first pass if contamination surfaces, and never weld over paint, oil, or coolant residue.