Pot Metal Welding: Why It Fails and How to Fix It

The chrome hood ornament looked perfect until you flipped it over and saw the crack running through the mounting stud. You fired up the TIG, touched the arc to the metal, and watched a silver puddle collapse, spit, and harden into a brittle grey scab. That moment — the one where pot metal turns from a part into a puddle — is where most welding advice quietly stops being useful.

Pot Metal Welding: The Honest Answer Before You Strike an Arc

A 1930s lamp base lands on the bench wrapped in newspaper, cracked clean through the socket collar. The owner has a TIG welder, a free Saturday, and reasonable confidence in his hands. Ten minutes later the collar is a slumped grey crater, the tungsten is contaminated, and the shop smells like a galvanised bucket left on a campfire. Nothing about that outcome was a technique problem.

The honest answer, before you waste another tungsten: conventional fusion welding — TIG, MIG, stick, oxy-fuel — is almost never the right process for pot metal. Not because you lack skill, and not because you picked the wrong amperage. Because the physics refuses to cooperate. Zinc, the dominant ingredient, melts at about 419.5 C (787 F) and boils at roughly 907 C (1,665 F). A welding arc burns in the thousands of degrees Celsius, far beyond that boiling point. There is no comfortable window where the base metal sits liquid and the filler bridges the gap. The metal goes from solid to collapsing to vapour, and it takes your filler, your shielding gas, and your patience with it.

The surface makes it worse. Zinc oxide, the grey-white skin that forms on the outside of a casting, melts at roughly 1,975 C. So while the metal underneath turns to soup, the oxide layer on top stays stubbornly solid and blocks any chance of wetting or fusion. That is the whole story in two sentences, and it explains nearly every failed pot metal weld you will ever see.

The Vocabulary You Need Before Anything Else

Pot metal is a shop-floor term, not a specification. It comes from foundries ladling molten metal out of a pot, usually whatever scrap was cheap that week. Which is exactly why pot metal behaves so inconsistently: two parts that look identical on a swap meet table can be entirely different alloys with entirely different melting points and entirely different repair odds.

You will hear the same material called several things, and they are not interchangeable:

• Die-cast zinc — the engineering description for zinc alloy parts made in a die, the most common construction for trim, housings and brackets.
• Zamak — the named zinc-aluminium family (Zamak 3, 5, 2 and 7) with controlled compositions. These are the good case, and they are not the norm in antique castings.
• White metal — a vague trade term for low-melting alloys, often tin or lead based, used in bearings, badges and cheap decorative trim.
• Monkey metal and muck metal — the blunt British and Australian shop terms for exactly the sort of unknown, impurity-laden casting that ruins your afternoon.

Welding, Brazing, Soldering: The Line That Explains Everything

The American Welding Society draws a clean boundary. Welding melts the base metal and the filler together into one pool. Brazing uses a filler that melts above 450 C (842 F) but below the melting point of the parent metal. Soldering uses a filler that melts below 450 C (842 F). Three processes, one rule: the parent metal stays solid.

Now hold that against a zinc casting that liquefies at 419.5 C. Even the soldering ceiling sits above the point where your part turns to liquid. That single number is why almost every successful pot metal repair is, technically, not welding at all — it is low-temperature soldering with a zinc- or tin-based filler, adhesive bonding, or a mechanical fix. Where a filler genuinely melts above 450 C and the casting survives, you are in the territory covered by torch brazing and air-acetylene fundamentals, and even there the margin is razor thin.

Who This Guide Is Written For

Most people who search for pot metal welding are not welders by trade. They restore vintage cars, jukeboxes, table lamps, guitar hardware, outboard fittings and chrome trim. They have a cracked casting, a parts supply that dried up in 1974, and a strong preference not to spend six hundred dollars on a replacement. Mixed in are professional shop welders handed a mystery casting by a customer who says “it’s just aluminium, right?” — and who need a defensible answer fast. If you are choosing consumables for a job like this, the same logic that governs filler metal selection applies: match the filler to the parent, never the other way around.

What’s Ahead — and What “Fixed” Actually Means

This article covers what pot metal really is, why fusion welding fails at the metallurgical level, and a full comparison of every realistic option: TIG brazing, low-temperature torch work, soldering, epoxy, mechanical reinforcement, replacement and recasting. It walks through a step-by-step repair procedure for cosmetic and lightly loaded castings, treats fume safety as an engineering topic rather than a disclaimer, and closes on cost and when to walk away.

Set your expectations honestly. Sometimes the right answer is replace, reinforce or recast — and a clean, well-reasoned refusal to weld is a professional outcome, not a defeat. Pot metal repair is very achievable once you stop trying to fuse it and start thinking like a restorer.

What Pot Metal Really Is: Composition, History, and Why the Name Matters

Before you can fix a mystery casting, you have to know what you are holding. The uncomfortable truth is that “pot metal” is less a material specification than a confession. The name comes from old foundry practice: a worker ladled molten metal from a communal pot holding whatever scrap was on the floor that day. Broken fittings, rejected castings, sprues, bearing metal, maybe a few lead weights. It was cheap, it poured well, and it took a polish. Nobody wrote down the recipe because nobody cared.

That history is the root of every repair problem ahead. A modern die caster works to a certifiable alloy. An antique lamp base or car mascot does not. The general description of pot metal captures it: a low-melting, inexpensive alloy of imprecise composition, used because it casts easily and machines nicely, not because anyone engineered it for strength or weldability.

How the Part Was Made Matters as Much as What It Is Made Of

Three casting routes dominate, and each leaves a different fingerprint on the metal.

• Sand and gravity casting: molten metal is poured into a mould and fills it under gravity. Cooling is slow, grain structure is coarse, and gas has a chance to escape. These parts are usually thicker and more forgiving.
• Hot-chamber die casting: molten zinc alloy is injected into a steel die under high pressure, very fast. This is how most trim, badges, handles and housings were made. It produces thin walls, crisp detail and excellent surface finish, and it traps air and combustion gas from the die lubricant inside the part.

That trapped gas matters later. Apply heat and those pressurised pockets expand, which is why a die casting can spit, bubble and collapse instead of flowing. The North American Die Casting Association publishes process and alloy guidance if you want the engineering-level version.

The Alloy Families You Are Likely to Meet

The best case is a modern zinc-aluminum die casting alloy from the Zamak family. Zamak names are built from the alloying elements: zinc, aluminum, magnesium and copper. Zamak 3 is roughly 96 percent zinc, 4 percent aluminum and 0.04 percent magnesium, with a melting range around 381 to 387 C (717 to 729 F). Zamak 5 adds about 1 percent copper, raising strength and hardness at some cost to dimensional stability and castability. Zamak 2 and Zamak 7 round out the family, with higher copper or a nickel addition respectively.

Antique parts are rarely that clean. Older zinc castings often carry lead, tin and cadmium picked up from scrap, and some were deliberately alloyed with lead for machinability. You will also meet aluminum-rich pot metals that behave like a dirty aluminum casting, and tin/lead white metal used for bearings, badges and cheap decorative trim, which melts low enough to soften in a hot car.

Likely alloy	Typical composition	Approx. melting range	Repair outlook
Zamak 3	~96% Zn, 4% Al, 0.04% Mg	381–387 C (717–729 F)	Best case; low-temp brazing possible
Zamak 5	Zamak 3 plus ~1% Cu	Approx. 380–386 C	Similar, slightly harder and stronger
Zamak 2 / 7	Higher Cu / Ni-bearing variants	Approx. 379–390 C	Specialist territory
Old leaded zinc	Zn with Pb, Sn, Cd impurities	Varies; often below 380 C	Poor; toxic fume risk
Aluminum-rich pot metal	Al with Zn, Si, Cu scrap	Approx. 540–650 C	Different process entirely
Tin/lead white metal	Sn-Pb alloys	Approx. 180–240 C	Soldering only; very weak

Zinc Pest: Why Some Old Castings Are Already Dying

Zinc pest is intergranular corrosion driven by lead, tin and cadmium impurities at the grain boundaries. It is well documented in antique die-cast toys, model trains and hardware. The part swells, warps, cracks in a fine network, and eventually crumbles into grey flakes. Humidity accelerates it. Once a casting is actively pesting, no filler will save it, because the crack you see is only the symptom of a material failing throughout its section.

Practical Identification Before You Heat Anything

You will rarely get a definitive alloy ID in a home shop, but you can get close enough to decide.

• Weigh it in your hand. Zinc die casting alloy runs about 6.6 g/cm³, roughly 2.4 times the density of aluminum at about 2.7 g/cm³. A part that looks like aluminum but feels oddly heavy is a strong hint.
• Magnet test. It rules steel in or out, but says nothing about zinc versus aluminum. Do not stop here.
• File a hidden edge. Zinc is bright silver-grey, softer than aluminum, and tends to load the file rather than cut cleanly. If you normally work aluminum, our aluminum MIG welding guide shows how differently that metal behaves under heat.
• Look for white powder. Zinc corrosion products are chalky white, often in crevices and around old cracks.
• Hot-iron touch test. On an inconspicuous spot, touch a low-temperature soldering iron to the metal. Zinc softens and melts far sooner than aluminum. Work outdoors or with extraction, because you do not want to breathe fume from an unknown antique alloy.

That last point is not paranoia. Antique pot metal may contain lead and cadmium, both serious fume hazards when heated, and the OSHA cadmium standard exists because these exposures are easy to underestimate in small shops.

Identify what you can, assume the worst about what you cannot, and treat every unknown casting as low-melting, gassy and possibly toxic. That mindset is half the repair already.

Why Welding Pot Metal Fails: The Metallurgy Behind the Mess

Every failed pot metal weld announces itself before you understand it. The puddle caves in, the arc spits molten metal at your glove, cracks appear as the part cools, and a grey-black scum floats to the surface. None of that is a technique problem. It is chemistry and physics doing exactly what the numbers predict, and once you can name the mechanism, the spitting stops looking random.

Here they are, roughly in the order they ambush you.

1. The Base Metal Melts Before the Filler Bridges the Joint

Zamak 3, the best-case alloy in antique work, begins to melt near 385 C and is fully liquid by about 390 C. ER70S-6 steel filler melts around 1,400 C; silicon bronze lands near 950 C. By the time your filler is ready to flow, the casting underneath has already lost its structure. You are not welding a joint — you are pouring filler into a hole that keeps widening. On thin trim, the part simply vanishes.

2. Zinc Boils at 907 C and Leaves in a Cloud

Zinc’s boiling point of about 907 C (1,665 F) is a couple of seconds of arc away, and a TIG arc runs well above 3,000 C. The metal does not merely melt; it vaporises, and the vapour expands violently as it escapes. That is the spitting you saw — expanding gas blowing molten metal out of the puddle. What remains is a spongy, undercut crater instead of a bead.

3. Zinc Oxide Melts at Roughly 1,975 C

Zinc loves oxygen. The oxide skin that forms instantly melts at around 1,975 C, far above the base metal’s own melting point, so you can never float it off. The metal beneath turns liquid and runs while the skin stays put. Argon and CO₂ do not remove it chemically, which is why mechanical and chemical cleaning before welding matters more than any machine setting. Filler that cannot wet clean metal cannot fuse, so it balls up and sits there like solder on a greasy pan.

What is happening	Approximate temperature
Zamak 3 base metal melts	385 C (725 F)
Zinc boils and vaporises	907 C (1,665 F)
Zinc oxide melts	1,975 C (3,587 F)
Typical TIG arc	above 3,000 C

4. Trapped Gas From the Die Casting Comes Back Out

Hot-chamber die casting forces liquid zinc into a steel die at speed, and that turbulence traps air and gas inside the part. It is a recognised limitation of the process, and one reason production die casters avoid weld repair on castings altogether. Apply heat and the trapped gas expands, opening porosity and blistering the surface. Your bead is porous before you finish it, and the surrounding skin can bubble like a pancake.

5. Thermal Expansion Runs More Than Twice Steel’s

Zinc alloys expand about 27 micrometres per metre per degree C; steel is roughly 12. Heat a small patch of a casting and the hot zone tries to grow while the cold metal around it refuses to move. The result is steep gradients, high restraint and cracking — often a fresh crack a few millimetres from the one you were repairing. It is the same heat-management problem behind distortion in cast aluminium wheel repair, except the melting point here is 275 C lower and the margin for error is gone.

6. Thin Walls, Bosses and Inserts Steal Your Heat

Die castings are designed for stiffness at minimum weight: 2 mm walls, thick mounting bosses, cast-in brass inserts. Those heavy sections act as heat sinks, pulling energy away from the joint while you chase the arc. You turn up the amperage to compensate, which pushes the thin section past its melting point. The outcome is lack of fusion on the thick side and a hole on the thin side.

7. Impurities Cause Hot Shortness and Grain Growth

Recycled zinc is rarely clean. Lead, tin, cadmium and iron are the classic embrittling impurities, and they migrate to grain boundaries where they form low-melting films. Heat the part and those boundaries open before the bulk metal reaches its melting point — hot shortness. Even a bead that looks sound can crack on cooling because the grain structure around it has been coarsened and weakened.

8. The Crack Was a Symptom, Not the Disease

Most cracks in old pot metal begin with intergranular corrosion, the zinc pest driven by those same impurities. The metal flanking the crack is already porous, weak and full of micro-cracks. Welding it is like stitching a wound in rotten cloth: the tear simply migrates to the next weak point. That is why a repaired bracket fails again three days later, a centimetre away.

Why Cleaning, Grinding and Better Technique Cannot Fix This

This is the part that stings. Every article about welding technique assumes the material stays solid while the filler melts, and that assumption is the whole ballgame. Pot metal breaks it at the phase level: the parent metal reaches liquid before the filler can wet it, then loses zinc to vapour, then traps its own oxide, then cracks as it contracts against cold metal.

Sharpening your arc, switching gas, or buying a more expensive tungsten changes none of that. You can grind a perfect V-groove and still watch the walls slump, because the problem is not joint preparation or travel angle. It is a phase-change problem, and the only real fixes are the ones that keep the parent metal below its melting point: brazing, soldering, adhesive bonding or mechanical repair. Those are the options the next section compares in detail.

Every Option Compared: TIG, MIG, Laser, Brazing, Solder, and Epoxy

Most people approach this question backwards. They pick a process first, then hope the part cooperates. The better habit is to ask two things before you touch a torch: what does this casting actually do, and how much of its original strength do I honestly need back?

Those two answers split every pot metal job into two completely different worlds. Cosmetic reconstruction — filling pits in a chrome trim ring, rebuilding a decorative edge on a lamp base, hiding a crack in a badge — tolerates a repair that only has a fraction of parent-metal strength, as long as it files, sands, and plates cleanly. Structural repair — a bracket, a hinge boss, a control arm, a mounting ear on marine hardware — needs load path continuity, and almost no pot metal process gives you that. A brazed or soldered joint in pot metal should always be treated as a fraction of parent-metal strength, never an equivalent replacement.

The Full Comparison Table

Process	Base metal melts?	Heat input	Strength vs parent	Cosmetic / plateable	Skill & cost	Verdict
TIG fusion	Yes — that’s the problem	Very high, concentrated	Fail	Blooms, spits, ugly	High skill, high risk	Don’t
TIG brazing (silicon or aluminium bronze)	No, ideally	High but controllable	Partial on thick Zamak	Good if cleaned	Specialist, steep failure rate	Thick sections only
MIG	Yes	High	Fail	Poor	Low skill, still fails	Don’t
Stick (SMAW)	Yes	Very high	Fail	Very poor	Low skill, worst fit	No
Oxy-fuel fusion	Yes	Broad and slow	Fail	Slumps badly	Moderate	No
Laser / micro-plasma	Yes, microscopically	Extremely localised	Industrial-grade only	Excellent	Tool-room equipment	Not a garage process
Torch brazing, low-temp zinc rod	No	Low	Low to moderate	Good, files well	Beginner-friendly	Best general DIY route
Soldering, tin-based filler	No	Very low	Low	Very good	Easy	Cosmetic and sealing
Structural epoxy	No	None	Moderate, peel-sensitive	Needs fairing	Easy	Good for non-peel joints
Mechanical repair	No	None	Often the highest	Hidden if planned	Moderate	Underrated, frequently best
Replacement part	—	None	Original	Original	Easy, variable cost	Always compare this first
Recast or machine from solid	—	Furnace / mill	Original or better	Excellent	High cost	For rare, valuable parts

What TIG Brazing Actually Is

TIG brazing is not welding, and calling it that is where people get hurt. You strike an arc, but the arc is only there to melt the filler; the casting stays solid. Silicon bronze filler melts around 1,000 C — far below steel, yet still above zinc’s boiling point of 907 C, which is why the margin for error is razor thin. The technique works on thicker Zamak sections by running low amperage, travelling fast, and backing the joint with copper to pull heat away. It produces a decent-looking, machinable bead that can be plated. It also fails often, especially on thin walls, and it belongs in the hands of someone who has already scrapped a few castings learning it.

For anything approaching a cosmetic edge, a low-temperature torch with a zinc-based rod is more forgiving. Those rods typically melt somewhere between roughly 175 and 400 C, so the casting never comes close to its own melting range. If you are setting up a torch for this, our guide to air acetylene welding, torch basics and brazing covers flame adjustment and tip selection in more depth.

The “Welds Pot Metal Like Steel” Claim

You will find rods advertised with exactly that promise. Read the melting point on the label. Most are zinc or cadmium-bearing alloys that flow a few hundred degrees above boiling water. They are not welding consumables and they do not fuse the base metal; they are low-temperature brazing alloys with good marketing. That does not make them useless — many do a respectable job on cosmetic castings — but it does mean the claim is false, and it means you must read the safety data sheet.

Cadmium-bearing low-temperature alloys genuinely work well, and cadmium fume is a regulated carcinogen. OSHA’s cadmium standard treats it as a serious chronic hazard, and cadmium is restricted in many jurisdictions. For a home shop, choose cadmium-free.

Laser, Adhesives, and the Industrial Options

Laser welding of zinc die castings is real and is used industrially, but it depends on carefully controlled porosity and tightly fixtured parts. It is not a garage process; our overview of alpha laser welding systems and their safety requirements shows just how much equipment sits behind that capability. Structural epoxies are advertised with tensile strengths in the low thousands of psi, but real joint strength collapses under peel loading and depends almost entirely on surface preparation — abrasive blasting, not a wipe with solvent.

Matching the Process to the Job

• Pitted chrome trim or a decorative edge: low-temperature torch brazing, then solder, then plate.
• A cracked but lightly loaded boss: mechanical repair — a backing plate, pin, or threaded insert — reinforced with epoxy.
• A thick, non-critical casting you can fixture: TIG brazing, if you have the skill and a copper backing.
• Anything carrying load, heat, or pressure: replace, recast, or machine from billet.

The pattern is consistent. The lower the process temperature, the higher your odds — and the lower the strength you can claim. Pick the coolest process that still does the job, and be honest with yourself about which job you are doing.

A Realistic Step-by-Step Low-Temperature Repair

Everything up to this point has been diagnosis. This is the part where you actually fix something. The procedure below is written for cosmetic castings and lightly loaded parts: a lamp base, a badge, a trim ring, a hinge boss on a jukebox door. It works with a low-temperature zinc-based rod or a torch brazing filler. It is not a recipe for a control arm, and no amount of careful technique changes that. Read the whole sequence before you pick up a torch, because most of the steps that decide the outcome happen before any heat arrives.

Before You Heat: The Go/No-Go Check

Ask three questions and answer them honestly.

• Structural or cosmetic? If the part holds load, holds pressure, or sits below the waterline, stop here.
• Will the repair be visible? A visible repair on bright trim demands near-perfect blending, which is a different skill set from making a bracket hold.
• What does a replacement cost? If a used part is thirty dollars on a restorer forum, an hour of careful torch work rarely makes economic sense.

Then identify the alloy as far as you can: weight in the hand, magnet test, file colour, white corrosion powder in the crevices. If you cannot get close, assume the worst case and work cooler than you think you need to. Unknown alloy means unknown outcome.

The Sequence That Actually Works

1. Assess and commit. Decide cosmetic versus structural, note anything safety-critical, and write the decision down.
2. Identify the alloy. Compare density against aluminium, check for magnetism, and look at the file scratch colour. Die-cast zinc reads dull grey and heavy for its size.
3. Stop-drill the crack ends. A small bit drilled just past the visible tip of each crack arrests propagation, the same crack-arrest logic used across metal repair.
4. Open the joint. Cut a 60–90° V-groove with a carbide burr at low speed, or scarf thin sections back to a feathered edge. A tight butt joint in pot metal is a guaranteed failure.
5. Degrease properly. Acetone or a residue-free solvent, applied until the rag stops coming away grey, then let it flash off completely. Good metal cleaning and prep practice applies here just as it does to aluminium.
6. Clean to bright metal. Use a stainless brush kept exclusively for zinc, so you do not embed steel particles in the joint.
7. Apply flux sparingly. Zinc chloride fluxes are aggressive and genuinely effective on zinc, but the residues are corrosive and must come out afterwards. Understanding how aggressive flux behaves at temperature helps you use less, not more.
8. Protect the neighbours. Heat-sink paste on threads and inserts, wet rag or wet clay against adjacent plating. Plating damage is permanent.
9. Preheat the whole part. A heat gun, not a torch, moved constantly until the casting is evenly warm at roughly 150 to 200 °C.
10. Verify the temperature. A temperature crayon or infrared thermometer, every time. Zinc gives almost no colour warning before it slumps.
11. Choose the smallest torch tip that will still melt the filler, and keep the flame moving in small circles. The basics of flame control are covered in any good torch setup and brazing guide.
12. Tin the joint. Touch filler to the heated parent metal and let it wet out; do not stab at it with the flame.
13. Build up in short passes, letting each one solidify before adding the next. Long, hot runs are how castings collapse.
14. Cool slowly. Air cooling only. Quenching a zinc casting introduces cracking and distortion that shows up days later.
15. Neutralise and remove flux. Warm water with baking soda and a stiff brush, repeated until no residue remains in the pores.
16. Finish and inspect. Coarse then fine file, carbide burr at low speed, then magnification or dye penetrant to confirm the crack has not crept.

The Tack-and-Check Rhythm

Do not try to run a continuous bead. Work in a rhythm: tack, stop, look, tack again. Between tacks, watch the edges of the puddle for the moment the parent metal starts to glisten and slump. That is your cue to back off the heat immediately. The rhythm also lets you keep the whole part warm rather than putting all your heat into one square centimetre.

Heat Sinks: Friend and Enemy

A copper or aluminium backing plate pulls heat out of a thin section and stops it sagging. It also pulls heat away from the joint you are trying to fill, which can leave you with lack of fusion and a repair that peels off under thumb pressure. The fix is placement: back the thin edge you want to protect, not the joint you want to wet. Never clamp directly across the repair zone.

Working Near a Thin Cosmetic Edge

Feather the heat in from thicker material toward the edge, never start at the rim, and accept that the last two millimetres may need to be built with filler rather than fused. If the edge starts to round over, stop, let it cool, and come back. Chasing it hot only makes it worse.

Done in this order, a low-temperature repair on a cosmetic casting is a genuinely achievable job. Done out of order, heat first and clean later, it becomes the grey scab that started this article.

Safety First: Zinc Fume Fever, Cadmium, and the Rules That Apply

Every pot metal repair is a fume-generating event. You are heating a metal that is mostly zinc to within a few hundred degrees of its boiling point, often with a torch flame playing directly on the surface. Ventilation and respiratory protection are part of the job design, not paperwork you skim past. The hazards are well documented and the controls are not exotic, but most home-shop injuries here come from exposure you never notice at the time.

Metal Fume Fever: The “Flu” You Gave Yourself

OSHA names zinc oxide fume from zinc-enriched metals as the usual cause of metal fume fever in welding and cutting operations (OSHA shipbuilding and repair guidance). The symptoms read like a bad cold that arrives on a schedule:

• Fever, chills, muscle aches, headache
• Metallic taste in the mouth, dry cough, chest tightness

They typically appear four to twelve hours after exposure and usually clear within 24 to 48 hours (metal fume fever overview).

That delay is why so many people blame a virus. You finish the repair at 6 p.m., feel fine, and wake at 3 a.m. shaking. By the next afternoon you are mostly recovered, so nothing gets connected.

There is also the “Monday morning” pattern. Welders who work galvanized or zinc-bearing metal through the week and take weekends off often report symptoms on the first day back. It is not tolerance that faded. It is simply re-exposure after a gap. No lasting immunity develops, and the shop myth that you can “get used to zinc” is false and dangerous, because habituation to the symptoms does not mean habituation to the exposure.

Cadmium: The One That Doesn’t Wash Out

Old pot metal can contain cadmium, and so can some low-temperature “miracle” repair rods still sold for cast zinc. Cadmium and its compounds are classified as known human carcinogens, which changes the risk calculation completely. OSHA’s cadmium standard, 29 CFR 1910.1027, sets a permissible exposure limit of 5 micrograms per cubic metre as an 8-hour time-weighted average, with an action level of 2.5 micrograms per cubic metre. It also carries medical surveillance, hygiene and recordkeeping requirements once exposures approach those thresholds. Those limits are tiny. You cannot see, smell or taste the difference between a safe plume and one that exceeds them.

Chromium, Lead, and Manganese

Chrome-plated trim brings hexavalent chromium into the conversation. Grinding or heating plating can generate Cr(VI) fume, and 29 CFR 1910.1026 sets the same 5 micrograms per cubic metre PEL as an 8-hour TWA. Lead shows up in older alloys and in paint on lamp bases and automotive trim. Manganese appears in some steels and specialty fillers and is linked to neurological effects with chronic overexposure. A stainless brush or stainless filler on a zinc part adds another chromium source. The practical takeaway: assume the plume is a mixture, because it usually is.

Engineering Controls Come First

Respirators are the last line of defence, not the first. Move the fume before you filter it. Local exhaust ventilation at the source is the benchmark: a capture hood or extraction arm positioned within a few inches of the work, not across the room. For small bench jobs, a fume extraction arm or a torch-mounted nozzle does more than any mask. Dilution ventilation via open doors and fans helps, but it spreads the plume rather than removing it at the arc. Working outdoors is genuinely effective. Keeping your head out of the plume is free and often overlooked, as is the habit of not leaning over the part to watch the puddle.

Respirators: A Supplement, Not a Solution

If you generate fume indoors, a fit-tested half-face or full-face respirator with particulate filters is the minimum, and it must be fit-tested to be worth wearing. For longer jobs, a powered air-purifying respirator (PAPR) reduces breathing resistance and improves compliance, which matters more than the spec sheet. Confined spaces, tanks and anything with poor air exchange call for supplied air, full stop. Facial hair breaks the seal, and a mask stored on a shelf for a year is not a control measure. If you also run a laser or micro-plasma setup, the fume and shielding considerations differ enough to warrant reading a dedicated safety guide such as Alpha Laser Welding: Systems, Uses, and Safety Guide before you start.

Fire, Housekeeping, and the Boring Details

Zinc dust from filing and grinding is combustible, and fine metal dust plus sparks is a classic ignition pair. Keep the bench clear, use a grounded metal bin for swarf, and never let a pile of filings accumulate under the work. Note that 29 CFR 1910.252 covers ventilation, personnel protection and fire prevention for welding generally, and it is the baseline any professional shop is measured against. Eye protection, FR clothing, leather gloves and a face shield belong on every job. If you can choose your gas and equipment setup, a supplier with proper safety documentation helps; the AllGas Welding Supply guide is a reasonable starting point.

Finally, if fume exposure is routine rather than occasional, stop guessing and monitor the air. Personal sampling tells you whether the controls work, which is information a respirator can never give you.

Preparation, Fixturing, and Finishing: Where Most Repairs Are Won or Lost

Ask ten restorers which low-temperature rod or paste they use and you will get ten answers and a small argument. Ask them how they cut the groove, held the part, and finished it afterward, and the answers converge fast. That is not a coincidence. On pot metal, the filler brand is the least important variable in the chain. Joint geometry, heat control, and post-repair finishing decide whether the repair survives a season or a decade.

Joint Geometry: Give the Filler Somewhere to Live

Zinc die castings do not tolerate a tight butt joint. A hairline gap gives the filler nowhere to key in, so it sits on the surface like a bead of dried glue and pops off the first time the part flexes. Instead:

• V-groove solid sections at roughly 60 to 90 degrees included angle, down to about two-thirds of the wall thickness. Leave the last third intact as a root so the part stays aligned.
• Scarf thin sections — a long, shallow taper on both faces rather than a deep narrow groove. Thin zinc walls have no heat capacity to spare, and a wide shallow scarf spreads the torch time instead of concentrating it.
• Skip the root gap. Unlike steel, pot metal has no penetration to chase. A gap simply lets filler run through and pool on the back side, which is where your cosmetic face used to be.
• Stop-drill both crack ends with a small bit before you touch a torch. A crack that keeps running under heat is not a joint, it is a moving target.

Fixturing Without Building a Heat Sink

Every clamp touching the casting is a heat thief. Small spring clamps and light C-clamps with copper or aluminium jaw pads work well, because they hold alignment without pulling heat out of the repair zone. Copper backing plates are your friend when you need to support a thin edge and pull heat away from a delicate flange — and your enemy when you actually want the joint to reach temperature. Know which you need before you clamp up.

For insulation, ceramic fibre blanket or a folded fire blanket under and around the part keeps heat where you put it, so you can preheat gently instead of chasing a cold casting with a hot torch. A simple jig of fire bricks lets the part grow and shrink while staying in alignment. Rigid fixturing that forbids thermal movement generates its own cracks.

Balance heat from both sides of a symmetric part, work in short passes, and let the casting cool between them. Light peening with a small hammer on a built-up area relieves contraction stress, but go easy — soft zinc moves more than you expect.

Finishing: Where Restorers Get Caught

Files load instantly with soft zinc, so keep a file card handy and work coarse to fine. Run carbide burrs at reduced speed so they cut rather than smear the alloy across the repair. Smearing looks fine wet and terrible under plating.

Paint is a chemistry problem. Zinc forms a stable oxide skin almost immediately, so ordinary topcoat alone rarely bonds; an epoxy or etch primer formulated for zinc is the reliable base. Clean and abrade immediately before priming, not the night before.

Plating is where projects stall. Plating baths are extremely sensitive to contamination, and shops will refuse parts containing lead, cadmium, or a high-zinc repair alloy rather than risk the chemistry of the whole tank. The conventional route for bright trim is a copper strike, then nickel, then decorative chromium. Seal porosity first and tell the plater exactly what you used.

Threads deserve their own plan. Chased threads in old zinc crumble; a threaded insert or a drilled-and-tapped oversize fastener is the durable answer. And remember galvanic reality: zinc is anodic to steel and brass, so a zinc casting bolted to a steel bracket in wet service will corrode sacrificially. Isolate the joint with a barrier washer and grease. Good surface prep habits carry over from aluminium work, and the same discipline applies here.

Short Note: Photograph the repair, note the filler alloy, flux, and cure temperatures used, and tuck a label inside the part. The next owner — or the next restorer — will thank you, especially before they put it near a plating tank or an oven.

Mistakes, Troubleshooting, and When to Walk Away

Most pot metal repairs do not fail at the torch. They fail in planning, or three weeks later when the plating lifts. These ten mistakes account for most of the wrecked castings that cross a restoration bench.

1. Treating pot metal as one known material. Pot metal is a scrap-derived family, not a specification. Zamak, leaded zinc and white metal behave differently, so an unknown alloy means an unknown outcome.
2. Going straight to the TIG torch. Fusion welding is usually the wrong process for die-cast zinc; the base metal collapses before a sound weld can form.
3. Overheating because zinc gives no colour warning. Unlike steel, zinc alloys do not show a reliable red glow before they slump, so the first clear signal is often a hole.
4. Skipping flux or choosing the wrong one. Zinc oxide will not wet out under shielding gas alone. A suitable aggressive flux is part of the process, not an optional extra.
5. Using too large a torch tip. Excess heat spreads into thin walls and bosses before the joint is ready, which makes control almost impossible.
6. Leaving flux residue behind. Trapped flux keeps working after the job, causing corrosion, staining and plating failure.
7. Quenching the part. Fast cooling locks in stress and can crack both the repair and the surrounding casting.
8. Repairing a crack without stopping it. A crack that is not stop-drilled will usually keep propagating under service vibration, even when the visible portion is filled.
9. Expecting a cosmetic repair to carry load. Filler and adhesive are not a substitute for sound parent metal.
10. Finishing with a process that re-melts the repair. Powder coating and some paint cycles run hot enough to soften low-temperature filler.

When a repair goes wrong, the symptom usually points to a cause you can fix. Work through this table before adding more heat or filler.

Symptom	Likely cause	Corrective action
Filler balls up and refuses to wet	Oxide, missing or weak flux, or too little heat	Clean to bright metal, apply fresh zinc-compatible flux, preheat the whole area, and use a soft flame with the smallest tip that holds temperature.
Casting slumps or develops a hole	Overheating; zinc gave no colour warning	Stop and let it cool. Cut back damaged metal only if enough section remains, then restart with lower heat and broader preheat.
Repair cracks again within days	Crack tips not stop-drilled, thermal stress, or contamination	Drill the crack ends, re-groove the joint, remove the old filler, and cool slowly.
Black scum or grey sludge forms	Oxidised zinc and flux residue	Remove it mechanically, neutralise and rinse, then restart with the correct flux.
Bubbles and porosity appear in the bead	Trapped die-cast gas, moisture, or oil	Preheat gently, clean thoroughly, and accept that some gassing is inherent to die castings.
Part warps or a boss moves	Uneven heat and inadequate fixturing	Re-fixture with allowance for expansion, use balanced heat, and insulate. If alignment is lost, replace the part.
Plating blisters or peels over the repair	Flux residue, porosity, or incompatible filler	Strip the finish, remove residue and porous metal, and replate only after testing the repair.
Joint pulls apart under light hand pressure	Lack of fusion, wrong filler, or no wetting	Cut it out and redo the joint properly. A patch over an unfused joint is not a repair.
White powder returns around the repair	Active intergranular corrosion	Replace the part. Cleaning treats the symptom; the corrosion can continue inside the casting.
An adjacent original crack appears	Stress redistribution or continuing zinc pest	Stop and reassess. Replacement is usually the honest answer.

When to Walk Away

Some parts should never see a torch in a general shop. Walk away from steering, suspension, brake, seatbelt and lifting components; pressure-containing castings; marine hardware below the waterline; and anything whose failure could injure someone.

Safety-critical automotive castings should not be repaired by amateur brazing or adhesive bonding under any circumstances. The same caution applies to cast aluminum wheel repair, which is why even a more weldable material has strict limits. A filler that holds a lamp base is not a suspension component.

Pot metal is also not cast iron. Cast iron repair techniques do not transfer because the melting points, graphite structure and thermal behaviour are entirely different. Advice that works on a manifold can destroy a zinc die casting.

For a professional shop, liability is part of the decision. Check with your insurer before accepting a repair that could later fail on a customer’s vehicle, and put the refusal in writing. OSHA welding safety guidance covers fume control; it does not certify a structural repair.

Replacement and Recasting Alternatives

When repair is not sensible, three routes usually beat a second attempt: a used or new-old-stock part from a specialist restorer, CNC machining from billet for simple shapes, and a 3D-printed pattern for sand casting when the shape is complex. Lead time and cost vary, but replacement, used parts and machining from solid are frequently cheaper than a repair that fails twice. For rare trim, a machined or cast replica can also be more dimensionally stable than a patched original.

Marine hardware below the waterline deserves the same discipline as aluminum boat welding: if the part keeps water out or holds structure together, replacement beats hope.

Cost, Value, and a Practical Playbook for Pot Metal Repairs

By now the shape of the problem should be clear. Pot metal is a family of low-melting zinc castings rather than a grade you can look up in a handbook, and the torch is almost always the wrong first tool. What remains is the decision every restorer eventually faces: repair it, replace it, or hand it to someone who does this every week. Price matters, but it is only part of the maths.

Here is how the four realistic routes compare. Treat the figures as rough bands, not quotations, because labour rates and plating costs swing wildly between regions.

Route	Approximate cost	Time	Risk	Cosmetic result	Best fit
DIY low-temperature repair	Kit, flux and consumables, typically far less than one professional plating job	An evening to a weekend	Medium — heat damage happens fast and gives no warning	Fine on hidden or painted areas; visible chrome still needs plating	Cosmetic castings, lightly loaded brackets, parts you can afford to lose
Professional restoration specialist	Hourly labour plus plating; on common production parts the total often exceeds the part’s value	Weeks to months	Low, if the shop genuinely specialises	Excellent, including show chrome	Rare or irreplaceable trim, anything prominent on a restored vehicle
New old stock or used replacement	Wide range, from swap-meet bargains to collector pricing	Days to weeks of hunting	Low	As good as the donor part	Anything still circulating through the parts network
Reverse engineering, CNC or recasting	Highest single-unit cost, though it drops sharply if you cast a batch	Months	Medium — pattern and shrinkage errors	Excellent, and you can improve the original design	Obsolete load-bearing parts, or one part that unlocks several restorations

The Shop-Floor Playbook

Strip the whole article down and this is the checklist worth taping inside a cabinet door:

1. Identify before you heat. Magnet, weight, file colour, corrosion powder — and if it is actually aluminium, the rules change entirely. Our aluminum MIG welding guide covers that path properly.
2. Decide cosmetic or structural. Be honest. Structural means engineer, not workbench.
3. Choose the lowest-temperature process that can do the job — epoxy, solder, low-temperature rod, and only then brazing.
4. Control heat with even preheat, heat-sink paste and a small, moving flame.
5. Use the right flux, sparingly, and remove every trace with water and baking soda. The prep discipline in our aluminum cleaning guide transfers almost one-for-one.
6. Cool slowly. No quenching, no compressed air, no wet rag on a hot joint.
7. Finish with a process that will not undo the work — check powder cure temperature against filler melting point first.
8. Document it. Filler type, repair temperature and date, written inside the part or in the file, so the next technician or plater does not destroy it.

The Honest Conclusion

Pot metal welding in the literal sense is a losing proposition. Fusion welding asks a zinc die casting to behave like steel, and it cannot: the base metal melts before the filler bridges, zinc boils at 907 °C, the oxide refuses to flux off, and any gas trapped during die casting escapes the moment you add heat. But pot metal repair is very achievable once you stop trying to weld and start thinking like a restorer.

Everything in this guide hangs off a handful of ideas worth carrying forward: die-cast zinc and the Zamak family behave nothing like wrought metals; zinc pest has already weakened many antique castings before you touch them; metal fume fever is a real occupational illness, not a myth you can train away; low-temperature brazing and soldering win where fusion loses; flux residue left behind causes corrosion and plating failures; platers can refuse contaminated alloys; and powder coat curing can quietly reflow your repair. Note, too, that cast aluminium parts such as wheels genuinely do weld, which is why our alloy wheel welding guide reads so differently from this one.

So be encouraged, but stay honest with yourself. If the part is cosmetic, cheap and replaceable, a careful low-temperature repair is a perfectly respectable outcome. If it is rare, expensive, or carries any load that could hurt someone, the professional call — specialist, replacement, or honest refusal — is the skilled call. Knowing the difference is what separates a restorer from someone who owns a torch.

Frequently Asked Questions

Can you actually weld pot metal with a TIG welder?

Not in the usual fusion sense. A TIG arc reaches temperatures far above zinc’s boiling point of about 907 C, so the base metal vaporises and collapses before a proper weld pool can form. Some specialists use very low-amperage TIG brazing with silicon bronze on thicker Zamak castings, but it is a specialist technique with a high failure rate and is not a beginner repair.

What is pot metal made of?

There is no fixed recipe. Pot metal is a scrap-derived zinc-based alloy that may contain aluminium, copper, magnesium, lead, tin, cadmium or iron in unpredictable proportions. Modern die castings often use Zamak 3 or Zamak 5, which are far more consistent, but antique parts can be almost anything.

Is pot metal repair strong enough to use on a car?

For cosmetic trim, emblems, dash parts and non-load-bearing brackets, a well-executed low-temperature repair can last for years. For steering, suspension, brake, seatbelt, fuel or lifting components, do not repair it — replace it. Those parts are engineered as single-piece safety components and a brazed or glued joint is not an acceptable substitute.

What is metal fume fever and how long does it last?

Metal fume fever is a flu-like illness caused by inhaling metal oxide fumes, most commonly zinc oxide from welding or heating galvanised and zinc-rich metals. Symptoms such as fever, chills, aches, headache and a metallic taste usually begin four to twelve hours after exposure and resolve within 24 to 48 hours, but the exposure itself should be treated as a warning sign of inadequate ventilation.

Why does my repair crack again after a few days?

The usual causes are an unstopped crack tip, thermal stress from localised overheating, contamination that prevented proper bonding, or intergranular corrosion continuing in the surrounding metal. Stop-drilling the crack ends, slow cooling and thorough cleaning solve most repeat failures, but a casting suffering from advanced zinc pest will keep cracking no matter how good the repair is.

Can pot metal be chrome plated after repair?

Often yes, but the plater needs to know what you used. Lead, cadmium and some high-zinc repair alloys can contaminate plating baths or cause adhesion failures, and some shops will refuse such parts outright. A copper strike followed by nickel and decorative chrome is the usual route, and the repair must be fully blended and free of porosity first.

Can I powder coat a repaired pot metal part?

Check the cure temperature first. Many powders cure around 180 to 200 C, which is close to or above the melting range of common low-temperature pot metal fillers. If the cure temperature exceeds your filler’s melting point, choose a different finish such as a compatible primer and topcoat, or use a filler rated for the cure schedule.

How do I tell pot metal from aluminium or steel?

Pot metal is non-magnetic, noticeably heavy for its size at roughly 6.6 g/cm3 versus about 2.7 for aluminium, and produces white powdery corrosion rather than red rust. A file cut shows bright silver, and a soldering iron touched to a hidden area will sink in at a few hundred degrees. When in doubt, treat the part as zinc and keep the heat low.

Is cadmium still used in pot metal repair alloys?

Some legacy low-temperature alloys contain cadmium because it improves flow and wetting, and old castings themselves may contain cadmium. Cadmium fume is a recognised carcinogen and is tightly regulated, including under OSHA’s cadmium standard. For home shop work, choose cadmium-free fillers and treat any unknown antique casting as if it contains cadmium.

When should I give up and replace the part?

Walk away when the part is safety-critical, when the casting shows widespread zinc pest or crumbling edges, when the repair cost approaches the cost of a replacement or a machined duplicate, or when the part must survive vibration or load. A clean refusal and a replacement part is usually cheaper than a repair that fails after plating.